Your SlideShare is downloading. ×
0
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×
Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications

7,201

Published on

Meenakshi Nagarajan,Amit Sheth,Selvam Velmurugan, "Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications," Tutorial at WWW2011, Hyderabad, India, March 28, 2011. …

Meenakshi Nagarajan,Amit Sheth,Selvam Velmurugan, "Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications," Tutorial at WWW2011, Hyderabad, India, March 28, 2011.

More info at:
http://knoesis.org/library/resource.php?id=1030

http://www2011india.com/tutorialstr27.html

Published in: Education, Technology, Business
2 Comments
10 Likes
Statistics
Notes
No Downloads
Views
Total Views
7,201
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
186
Comments
2
Likes
10
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide
  • Got carried away with coverage and content – too much material for 3 hours – so the remaining content can be used as background
  • Got carried away with coverage and content – too much material for 3 hours – so the remaining content can be used as background
  • - We want to understand meaningful citizen sensor observation  social signals
  • Source for Stats
  • Many media companies use Facebook and Twitter as news-delivery platform. Many individuals rely on them as news source. News is increasingly social.
  • tweetmeme_url = 'http://www.readwriteweb.com/archives/this_is_what_a_tweet_looks_like.php'; tweetmeme_source = 'rww'; A tweet is filled with metadata - information about when it was sent, by who, using what Twitter application and so on
  • “isn't much of life's meaning found in the play between limits and the infinite?”
  • Link to media files, context about annotation, a special option to write reviews of movies, books, or links you're sharing. The ISBN of the book, a link to a preview of the movie and the number of stars in your rating could be included in the Tweet Annotations, Any way you can classify, describe, append or otherwise enrich a Tweet with words or numbers can be included in Annotations.
  • Interest level:(Based on Description info, lists and fav. tweets)
  • Semantic metadata, relationships: Inferred?
  • Structure Level MetadataCommunity Size - Showing scale: global vs. localCommunity growth rate - Popularity estimation for a topicLargest Strongly Connected Component size - Measuring Reachability in the directed graphNo. of Weakly Connected Components & Max. size - distribution of pre-existing network connections (follower-followee) - Showing Nature: loose vs. compactAverage Degree of Separation - How many hops between two authorsClustering Coefficient - Showing the likelihood of associationRelationship Level MetadataType of Relationship- topic/content (based on Retweet, Entity etc.)- follower/followee (based on structure)Relationship strength- Strong vs. Weak ties based on activity/ communication between users - % tie strengthUser Homophily [Homophily (i.e., "love of the same") is the tendency of individuals to associate and bond with similar others] based on certain characteristic (e.g., Location, interest etc.)% of users showing similar behaviorReciprocity: mutual relationship- % of users following back their followersActive Community/ Ties- How active is the communication between users or how active are the relationship ties - Average of tie strength based on activity
  • Pat Hayes
  • Add some examples of how people store such semantic metadata…When you put social data as LOD, talk about technologies -
  • Building on foundations of Statistical Natural Language ProcessingInformation ExtractionSemantic Web/ Knowledge RepresentationWe will talk about key issues in extracting metadata from Informal Text and how it varies from what has been done in more well-structured text like news articles etc.
  • Social Media text is informal for various reasons.. Read red points
  • Recently two researchers came up with a score to formalize the contextual nature of text and therefore the formality of text. More the available context, more formal the textWe used the same score on SM text and found that …---Score is too limited and probably outdated– does not consider full sentences/structure, does not consider links– similarly network related score would be good to have
  • What the two tasks look like in terms of outputs they produce
  • For two types of NE movie and music over two types of SM textUsing those cues
  • Focus only on one row at a time Cultural entity defn in next slide
  • What makes cultural entity extraction more difficult
  • There are two flavors to the Cultural entity recognition problemWhere same entity appears in multiple senses in the same domainWhere same entity appears in multiple senses in different domains
  • Focusing on the first flavor
  • Same song occurs as multiple instances in Music Brainz (knowledge base)
  • Sample real world constraints hard-coded – this work was an experiment into scoping using real-world constraints
  • As you chop away the domain model you accuracy increases…
  • This is an application of the NER work
  • Conclusion in RED
  • We have come a long way but still room for improvement
  • Fact can be proven, opinion cannot. An opinion is normally a subjective statement that bases on people's thoughts, feelings and understandings.
  • Social media serves as a platform for people to speak their mind more freely, which lead to a growing volume of opinionated data that can be used by:  (1) individuals for suggestion and recommendation(2) companies and organizations for marketing strategies and other decision making process(3) government for monitoring social phenomenons, being aware of potential dangerous situations, etc.
  • For the task of classification, supervised learning or unsupervised learning techniques can be used. For the review-like data, e.g., movie review or product review, it's easy to get training data from website like imdb or Amazon. The sentiment classification is different from traditional topic classification since they have different features involved.lexicon-based approach: first creating a sentiment lexicon, and then determining the polarity of a text via some function based on the positive and negative clues within the text, as determined by the lexicon. The idea of bootstrapping is to use the output of an available initial classifier to create labeled data, to which a supervised learning algorithm may be applied. The task of extracting the opinion/holder/target is similar to the traditional information extraction task. The difference is that for this task, the relations between opinion and opinion target are considered important.E.g. proximity, the opinion expression is assumed to be closed to the opinion target in the text. Based on this assumption, if the opinion target is given, then the nearby adjectives can be extracted as opinion candidates.Other possible ways to model the relations between opinion and opinion target include: syntactic dependency, co-occurrence, or manually prepared patterns/rules 
  •  In this paper, The authors connect measures of public opinion measured from polls with sentiment measured from tweets. They find that a relatively simple sentiment detector based on Tweets replicates consumer confidence and presidential job approval polls.The results highlight the potential of text streams as a substitute and supplement for traditional polling. Positive and negative words are defined by the subjectivity lexicon from OpinionFinder,a word list containing about 1,600 and 1,200 words marked as positive and negative, respectively (Wilson, Wiebe, and Hoffmann, 2005)A message is defined as positive if it contains any positive word, and negative if it contains any negative word. (This allows for messages to be both positive and negative.)
  • In this paper, the authors demonstrate how social mediacontent can be used to predict real-world outcomes. In particular, they use tweets to forecast box-officerevenues for movies. The results show that the prediction model using the rate at which tweets are created about a movie outperforms the market-based methods. And the sentiments present in tweets about a movie can be used to improve the prediction. The intuition is that a movie that has far more positive than negative tweets is likely to be successful. For the task of sentiment classification of tweets, they use a supervised classifier "DynamicLMClassifier" from LingPipe.Each tweet in the training set is labeled as positive, negative or neutral by workers from Amazon Mechanical Turk.The classifier is trained using the n-gram model. In their work, they use n=8.  they find that the sentiments do provide improvements, although they are not as important as the rate of tweets themselves 
  • One of the most attractive advantages of unsupervised approaches is that they do not require for training data.Many sentiment analysis applications for social media content use simple lexicon-based method. However, for the problem of target-specific sentiment analysis, it doesn't work. Based on simple lexicon-based method which use a general sentiment lexicon containing positive/negative/neutral words in the general sense, (1) for the task of "find tweets containing positive opinions about a specific topic", such as a movie, the results will like the table shows. However, 2,3,5,6,7 don't contain opinions about the movie. (2) for the task of extract the opinion clues/expressions, the right answers should be like we show in the other picture. However, the simple  lexicon-based method might give all the words with orange color in the table.
  • We create a general subjective lexicon which contains subjective words in the general sense. This lexicon is created by extending the commonly used subjective lexicon to involve slangs learned from Urban Dictionary.This general lexicon is used for select sentiment units candidates. A bootstrapping method is used to learn domain-dependent sentiment clues from domain-specific corpus. Most of the current lexicons only contain words. We employ statistical models to find words, phrases and patterns which can be used as sentiment clues. Such as "must see", "want my money/time back", "don't miss it" in the movie domain.For the task of identifying opinions towards the given target, we use a syntactic rule-based method as well as proximity model. Since the informal language structures of tweets bring difficulties to the parser, our method just requires a very shallow syntactic parse of tweets.
  • Refs:http://en.wikipedia.org/wiki/Writing_stylehttp://en.wikipedia.org/wiki/Psychometrics
  • Metadata from Network Analysis:- Not sufficient to answer the above questions unless we consider context, and hence merge approach (Content + Network) is better
  • [Example scenario:- Buyer wants to buy a movie dvdMultiple influencers!!!- Key Influencers: Media experts- Peer Influencers: Hiscollegues (the people, buyer interacts face-to-face daily)- Social Influencers: His social circle ]Now how do we find out them?Link Analysis based on structure is not just sufficient ---- SOCIAL MEDIA IS HIGLY COMPLEXThat’s why we need additional context analysis in play- Popularity NOT =Influence! - We need to understand audience, their activity level and interest is of greater importanceHomophily (tendency to follow similar behavior) limits people's social worlds in a way that has powerful implications for the information they receive, the attitudes they form, and the interactions they experience.KLOUT: (http://klout.com/kscore)Reach :: Are your tweets interesting and informative enough to build an audience? How far has your content been spread across Twitter?Amplification:: Probability is the likelihood that your content will be acted upon
  • Multiple types of users - HOW DO WE FIND OUT THESE TYPES?Does external web (background knowledge) presence of a user tells us more than the limited context available in the network?
  • User engagement levels: applications in coordination activities Connecting the dots here with NGO initiatives (*presented by Selvam)
  • User engagement levels: applications in coordination activities Connecting the dots here with NGO initiatives (*presented by Selvam)- Just not limit to Active vs. Passive in general but be specific to topic and then say ‘active’/passive w.r.t. topic (e.g., active for ‘Biology info’ vs. passive for ‘comp. sci. info.’)
  • Connections/Relationships- Implicit content features
  • We want to achieve by Network Analysis for social media: - Graph Traversal --- for understanding reachability between people - Community Formation, sustainability for people
  • We want to analyze these Social networks for understanding various social science studies:- DiffusionHomophily (tendency to follow similar behavior) – based on certain characteristic (demographic, interest etc.) What makes dynamics to be diff. here (factors)
  • Authoritativenature of the poster or the volume of follower connections did not predict the re-tweet behavior associated with the tweets!
  • Spammers diverting their attention to social media sites.
  • This tweet was by Kenneth Cole at the time of Egypt Revolution. Though it uses a hashtag that was used to indicate a tweet on Egypt crisis (#Cairo), the link it has is not connected to Egypt crisis.
  • This Article was published on Guardian website in Feb 2010. In this article the Director of BBC Peter Horrocks states that the journalists should use social media as the primary source of Information. He took over the position of Director a week back.Now let us consider a scenario where a Journalist wants to follow social signals wants to analyze what news is stirring up today at a particulat location.There is a problem using this since there is Information Overload
  • This use case requires merging streaming data with background knowledge information (e.g. from DBpedia). Examples of ?category include category:Wi-Fi devices and category:Touchscreen portable media players amongst others. As a result, without having to elicit all products of interest as keywords to lter a stream, a user is able to leverage relationships in background knowledge to more effectively narrow down the stream of tweets to a subset of interest.
  • How news articles We collected the output of our system for healthcare topic for a time period. We also collected articles from the Nytimes for the same period and put this as the input for our Extraction pipeline. And plotted the occurrence of entities in tweets and in Nytimes articles. We found that both these co-ordinate very well. We then got the events occurred during the peaks of our time plot from timeline.com and nytimes.com and found this result.
  • Transcript

    • 1. Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web ApplicationsTutorial at WWW2011, Hyderabad, IndiaMarch 28, 2011<br />1<br />
    • 2. Citizen Sensing<br />Overview, Social Signals, Enablers<br />Role of Social Media<br />Activism, Journalism, Business Intelligence, Global Development<br />Development-Centric Platforms<br />Beginnings, Architectures and Possibilities<br />Systematic Study of Social Media<br />Spatio-Temporal-Thematic + People-Content-Network Analysis<br />Trustworthiness in Social Media<br />Mobile Social Computing<br />Citizen Sensing @ Real-time<br />Research Application: Twitris<br />Conclusion & Future Work<br />2<br />Outline<br />
    • 3. Acknowledgements<br />3<br />
    • 4. Selvam Velmurugan <br />(Kiirti, eMoksha NGOs)<br />Meena Nagarajan<br />(Content Analysis)<br />Hemant Purohit<br />(People & Network analysis)<br />AmitSheth<br />(Semantic Web)<br />Ashutosh Jadhav<br />(Event Analysis)<br />Lu Chen<br />(Sentiment Analysis)<br />Pramod Anantharam<br />(Social & Sensor web)<br />Pavan Kapanipathi<br />(Real Time Web)<br />
    • 5. Preliminaries<br />Tutorial description: http://www2011india.com/tutorialstr27.html and http://knoesis.org/library/resource.php?id=1030<br />Lots of breadth: many examples, some depth: few algorithms, mainly to convey insights<br />Twitter > Myspace/Facebook > SMS<br />Each has different reach/focus/importance<br />Given the time, only parts will be covered today! Citations, further reading at bottom and at the end<br />Images belong to their copyright holders. Copyright info. for images, where available are at the end.<br />5<br />
    • 6. Aim<br />What are research opportunities and technical challenges in gaining insights and use of social media content (esp. citizen sensing)?<br />Provide a structure to a vast array of issues<br />Breath, not depth<br />6<br />
    • 7. Citizen Sensing<br />Overview, Social Signals, Enablers<br />Role of Social Media<br />Activism, Journalism, Business Intelligence, Global Development<br />Development-Centric Platforms<br />Beginnings, Architectures and Possibilities<br />Systematic Study of Social Media<br />Spatio-Temporal-Thematic + People-Content-Network Analysis<br />Trustworthiness in Social Media<br />Mobile Social Computing<br />Citizen Sensing @ Real-time<br />Research Application: Twitris<br />Conclusion & Future Work<br />7<br />Outline<br />
    • 8. Citizen Sensing<br />Common person (citizens of Internet) is able to use Web2.0 and social networks <br />The human centric activity** of observing, reporting, disseminating information (facts, opinions, views) via text, audio, video and built in device sensor (and smart devices)<br /> ** direct/indirect, collective/individual <br />Human-in-the-loop (participatory) sensing <br /> + Web 2.0 + Mobile computing<br />= Emergence of <br /> Citizen-Sensor networks<br />Image: http://bit.ly/hmZe428<br /> A. Sheth, 'Citizen Sensing, Social Signals, and Enriching Human Experience', IEEE Internet Computing, July/August 2009, pp. 80-85.<br />
    • 9. Understanding meaningful citizen sensor observations<br />Social Signal Processing: Aggregation, Enhancement, Analysis, Visualization, and Interpretation<br />Citizen-Sensor network:<br />Immense potential to disseminate social signals quickly and in real-time<br />9<br />Social Signals<br /> A. Sheth, 'Citizen Sensing, Social Signals, and Enriching Human Experience', IEEE Internet Computing, July/August 2009, pp. 80-85.<br />Image:http://bit.ly/gWHSjD<br />
    • 10. <ul><li> Mobile Platforms Hit Critical Mass, Over 5 billions users
    • 11. 1+B with internet connected mobile devices (2010)
    • 12. Smartphones> Notebooks + Netbooks (2010E)
    • 13. 500K+ mobile phone applications
    • 14. 74% of mobile phone users (2.4B) worldwide used SMS (2007)</li></ul>Mobile device might qualify as humankind's primary tool<br />Redefines the way we engage with people, information, etc.<br />Enablers: Mobile Devices<br />& Ubiquitous Connectivity<br />Mobile is Global<br />Ubiquity, 24x7<br />Built in sensors<br />environmental, biometric/biomedical,...<br />10<br />
    • 15. Enablers: Web 2.0 & Social Media<br />500M+ Facebook Users<br />100M+ Twitter users, 85M+ tweets/day<br />Internet Users: 1.8 Bln<br />Large variety of social media and traditional media interact, creating potent mixture <br />11<br />Types of UGC: <br /><ul><li> Twitter (text/microblogs)
    • 16. Facebook (multimedia)
    • 17. YouTube (videos)
    • 18. Flicker (images)
    • 19. Blogs (text)
    • 20. Ping(Social network for music) </li></ul>Image: http://bit.ly/euLETT<br />
    • 21. Citizen Sensing<br />Overview, Social Signals, Enablers<br />Role of Social Media (important classes of applications)<br />Activism, Journalism, Business Intelligence, Global Development<br />Development-Centric Platforms<br />Beginnings, Architectures and Possibilities<br />Systematic Study of Social Media<br />Spatio-Temporal-Thematic + People-Content-Network Analysis<br />Trustworthiness in Social Media<br />Mobile Social Computing<br />Citizen Sensing @ Real-time<br />Research Application: Twitris<br />Conclusion & Future Work<br />12<br />Outline<br />
    • 22. Citizen Sensors in Action<br />Mumbai Terror Attack<br />Iran Election 2009<br />Haiti Earthquake 2010<br />US Healthcare Debate 2009<br />13<br />Image: http://huff.to/hp0OhA<br />
    • 23. Revolution 2.0<br /> Political/Social Activism<br />Ghonim, who has been a figurehead for the movement against the Egyptian government, told Blitzer “If you want to liberate a government, give them the internet.”<br /><ul><li> When Blitzer asked “Tunisia, then Egypt, what’s next?,” </li></ul> Ghonim replied succinctly “Ask Facebook.”<br />http://cnn.com/video/?/video/world/2011/02/13/nr.social.media.revolution.cnn<br />http://cnn.com/video/?/video/tech/2011/02/11/barnett.egypt.social.media.cnn<br />Egyptian anti-government demonstrator sleeps on the pavement under spray paint that reads 'Al-Jazeera' and 'Facebook' at Cairo's Tahrir square on February 7, 2011. <br />http://www.cbsnews.com/stories/2011/02/15/eveningnews/main20032118.shtml<br />14<br />
    • 24. Citizen Journalism<br />15<br />Twitter Journalism<br />Images: http://bit.ly/9GVfPQ, <br />http://bit.ly/hmrTYV<br />
    • 25. News is increasingly Social<br />Social News<br />Social Media and Global Media are inter-twined.<br />16<br />
    • 26. Business Intelligence: <br />Trend Spotting, Forecasting, Brand Tracking, Targeted Advertising<br />Sysomos(http://www.sysomos.com/)  - Business intelligence by engaging, measuring and understanding activities in Social Media<br />Trendspotting(http://trendspotting.com) <br /> - Detecting, analyzingandevaluating trends for business.<br />Simplify(http://simplify360.com/)<br /> - A collaborativeplatform to monitor, measureandengage<br />customersusing Social Media.<br />Shoutlet(http://www.shoutlet.com/)<br /> - Managing social media marketing communication using a single<br />platform.<br /> Reputation.com(http://www.reputationdefender.com/)<br /> - Preserves privacyanddefendsreputationbyprotectingattacks<br />onpersonalinformation.<br />Image: http://bit.ly/eAebBb<br />17<br />
    • 27. Social Development <br />(Education, Health, eGov)<br />LiveMocha (http://www.livemocha.com/)<br />Online Language learning tool with social engagement <br />         - bridging the gap!!<br />Soliya(http://www.soliya.net/)<br />Dialogue between students from diverse backgrounds<br /> across the globe using latest multimedia technologies<br />ProjectEinstein <br />(http://digital-democracy.org/what-we-do/programs/)<br />A photography-based digital penpal program connecting youths in refugee camps to the world<br />PatientsLikeMe<br /> (http://mashable.com/2010/07/13/social-media-health-trends/)<br /> -   Facilitates sharing of health profiles, finding patients with similar <br /> ailments, and learn from discussions.<br />TrialX(http://trialx.com/) <br />- Finding clinical trials of new treatments and connecting with clinical trial investigators.<br />18<br />Image: http://bit.ly/ayyjlU<br />
    • 28. Citizen Sensing<br />Overview, Social Signals, Enablers<br />Role of Social Media<br />Activism, Journalism, Business Intelligence, Global Development<br />Development-Centric Platforms<br />Beginnings, Architectures and Possibilities<br />Systematic Study of Social Media<br />Spatio-Temporal-Thematic + People-Content-Network Analysis<br />Trustworthiness in Social Media<br />Mobile Social Computing<br />Citizen Sensing @ Real-time<br />Research Application: Twitris<br />Conclusion & Future Work<br />19<br />Outline<br />
    • 29. Collaboration<br />We “simply do not have enough genes to program the brain fully in advance,” we must work together, extending and supporting our own intelligence with “social prosthetic” systems that make up for our missing cognitive and emotional capacities: “Evolution has allowed our brains to be configured during development so that we are ‘plug compatible’ with other humans, so that others can help us extend ourselves.”<br />- Harvard "Group Brain Project"<br />20<br />
    • 30. Beginnings<br />Open Source <br />Linux, Apache<br />Social Networks<br />FaceBook, Twitter, MySpace<br />Crowd Sourcing<br />Wikipedia, Kiva, Ushahidi, Kiirti, SwiftRiver, Sahana<br />Collaborative Governance<br />Peer-to-Patent, e-Demogracia<br />21<br />
    • 31. Popular Initiatives<br />FaceBook + Twitter<br />Iran post-election protests<br />Tunisia and Egypt uprisings  <br />Ushahidi<br />Kenyan post-election violence<br />India, Lebanon, Afghanistan, and Sudan elections<br />Haiti Earthquake<br />Pakistan Floods<br />Kiirti<br />BBMP election monitoring<br />Bangalore AutoWatch <br />22<br />
    • 32. FixOurCity - Chennai<br /> Built on top of FixMyCity open-source codebase<br /> Stage I<br /> Report by Area/Ward and Street<br /> Integration with Google Map<br />Displays Ward member name/contact details<br />Select category of issue, description and severity<br />Confirmation through email to avoid misuse<br /> Stage II/III<br /> Normalize incoming reports to official wards and categories<br />Integration with Corporation website to allow auto-forwarding and updating of reports<br />23<br />
    • 33. Ushahidi<br />Information Collection: SMS (FrontlineSMS, Clickatell), <br /> Email, Web<br /> Visualization/Interactive Mapping: Timeline, Category, <br /> Geo-spatial<br /> Alerts: Geo-spatial<br /> Admin: User Management, Report Moderation / Creation, <br /> Site Statistics<br />24<br />
    • 34. SwiftRiver<br />Filtering and verification of real-time data from channels like Twitter, SMS, Email and RSS feeds. Offers organizations an easy way to apply semantic analysis and verification algorithms to different sources of information. <br />Speed up the process of managing real-time data streams (email, web, sms, twitter)<br />Add elusive context (location, historical data) and history (reputation of sources) to online research<br />Offer a dashboard for monitoring multiple channels of information<br />Offer advanced aggregation and analytic tools on or offline<br />Give the user control over advance curation tools and filter<br />25<br />
    • 35. SwiftRiver Architecture - I<br />26<br />
    • 36. SwiftRiver Architecture - II<br />27<br />
    • 37. Free and Open Source Disaster Management system. A web based collaboration tool that addresses the common coordination problems during a disaster between Government, the civil society (NGOs) and the victims themselves.<br />Sahana<br />
    • 38. Mapping - Situation Awareness & Geospatial Analysis.<br />Messaging - Sends & Receives Alerts via Email & SMS.<br />Document Library - A library of digital resources, such as Photos & Office documents.<br />Missing Persons Registry: Report and Search for Missing Persons.<br />Disaster Victim Identification<br />Requests Management: Tracks requests for aid and matches them against donors who have pledged aid.<br />Shelter Registry - Tracks the location, distribution, capacity and breakdown of victims in Shelters<br />Hospital Management System - Hospitals can share information on resources & needs.<br />Organization Registry - "Who is doing What & Where". Allows relief agencies to coordinate their activities.<br />Ticketing - Master Message Log to process incoming reports & requests.<br />Delphi Decision Maker - Supports the decision making of large groups of Experts<br />Sahana<br />
    • 39. Peer to Patent<br />Peer To Patent opens the patent examination process to public participation for the first time. It is an online system that aims to improve the quality of issued patents by enabling the public to supply the USPTO with information relevant to assessing the claims of pending patent applications.<br />30<br />
    • 40. http://www.peertopatent.org/video/p2p640/VideoPlayer.html<br />31<br />Peer to Patent - Video<br />
    • 41. Kiirti<br />Allows you to set up your own instance of the Ushahidi Platform without having to install it on your own web server. <br />Provides pre-integrated Voice and SMS reporting capabilities within India.<br />32<br />
    • 42. 33<br />Kiirti – Home Page<br />
    • 43. 34<br />Kiirti – User Interaction Flow<br />
    • 44. Kiirti - Flywheel of Engagement<br />35<br />
    • 45. Future Possibilities<br /> Online Dispute Resolution<br />30M+ pending cases in India's courts<br /> Public Policy Reviews<br />Crisis Management<br />Effective Local Governance<br />36<br />
    • 46. Challenges<br />Challenges<br />Information overload<br />Processing and de-duping messages<br />Accessibility (e.x. network congestion, access points, …)<br />Incorrect or partial data<br />Trustworthiness of source (e.x. influence, reputation, …)<br />Metadata extraction (e.x. geo data, name-entity, sentiment/opinion, …)<br />Collaboration<br />Policy discussions<br />Structure or hierarchy<br />
    • 47. Citizen Sensing<br />Overview, Social Signals, Enablers<br />Role of Social Media<br />Activism, Journalism, Business Intelligence, Global Development<br />Development-Centric Platforms<br />Beginnings, Architectures and Possibilities<br />Systematic Study of Social Media<br />Spatio-Temporal-Thematic + People-Content-Network Analysis<br />Trustworthiness in Social Media<br />Mobile Social Computing<br />Citizen Sensing @ Real-time<br />Research Application: Twitris<br />Conclusion & Future Work<br />38<br />Outline<br />
    • 48. Dimensions of Systematic Study of Social Media<br />Spatio - Temporal -Thematic+<br />People - Content - Network<br />39<br />
    • 49. Social Information<br />Processing<br />"Who says what, to whom,<br /> why, to what extent and with what effect?" [Laswell]<br />Network: Social structure emerges from the aggregate of relationships (ties)<br />People: poster identities, the active effort of accomplishing interaction<br />Content : studying the content of communication <br />40<br />
    • 50. Studying Online Human Social Dynamics<br />How does the (semantics or style of) content fit into the observations made about the network?<br /> <br />Often, the three-dimensional dynamic of people, content and link structure is what shapes the social dynamic. <br />41<br />Example: how does the topic of discussion, emotional charge of a <br />conversation, the presence of an expert and connections between participants; together explain information propagation in a social network?<br />Image: http://bit.ly/dFzjU2<br />
    • 51. Why People-Content-Network + Spatial-Temporal-Thematic metadata?(Example of Understanding Crisis Data)<br />42<br />
    • 52. Metadata/Annotations<br />Metadata: an organized way to study<br />Types <br />Creation/extraction and storage<br />Use <br />43<br />Image: http://www.biowisdom.com/tag/metadata/<br />
    • 53. Metadata Infrastructure: <br />Example for Tweet Annotation (mapped out tweet)<br />44<br />Image: http://rww.to/9zyoQa<br />
    • 54. 45<br />http://www.readwriteweb.com/archives what_twitter_annotations_mean.php<br />
    • 55. 46<br />
    • 56. `<br />People Metadata: <br />Variety of Self-expression Modes <br />on Multiple Social Media Platforms<br /> Explicit information from user profiles <br /> User Names, Pictures, Videos, Links, Demographic Information, Group memberships...<br />    <br /> Implicit information from user attention metadata<br />Page views, Facebook 'Likes', Comments; Twitter 'Follows', Retweets, Replies.. <br />47<br />
    • 57. People Metadata: Various Types<br />Identification<br />Interests<br />Activity<br />Network<br />48<br />
    • 58. People Metadata: Continued<br />49<br />
    • 59. People Metadata: Continued<br />Web Presence:<br />- User affiliations<br />- KLOUT Score – influence measure (www.klout.com)<br />50<br />
    • 60. Content Metadata<br />Content Independent metadata<br /> • date, location, author etc.<br />51<br />2. Content Dependent metadata<br />Direct content-based metadata <br />i. Explicit/Mentioned Content metadata<br />• named entities in content<br />ii. Implicit/Inferred Content Metadata<br />• related named entities from knowledge sources<br />b. Indirect content-based metadata (External metadata) <br />• context inferred from URLs in content (images, links to articles, FourSquarecheckins etc.)<br />V. Kashyap and A. Sheth, 'Semantic Heterogeneity in Global Information Systems: The Role of Metadata,<br /> Context and Ontologies,’ in Cooperative Information Systems: Current Trends and Directions, M. Papazoglou and G. Schlageter (Eds.), Academic Press, 1998, pp. 139-178.<br />
    • 61. Content Metadata: <br />Content Independent <br />For Tweets<br />Published date and time<br />Location (where tweet was generated from)<br />Tweet posting method (smart-phone, twitter.com, clients for twitter)<br />Author information<br />52<br /><ul><li>For SMS</li></ul>Publish date and time<br />Location (where SMS is generated)<br />Receiver (NGO, Government organization)<br />carrier information (available on request)<br />
    • 62. Content Metadata: <br />Content Dependent (Tweet)<br />53<br />Direct Content-based Metadata<br />Indirect content-based metadata (External metadata)<br />
    • 63. Content Metadata: <br />Content Dependent (SMS)<br />Direct Content-based Metadata<br />54<br />
    • 64. Network Metadata<br />Connections/Relationships matter! (foundation for the network)<br />55<br />
    • 65. Metadata: Creation, Extraction and Storage<br />56<br />
    • 66. Metadata Creation & Extraction<br />Extracted Metadata<br /> Directly visible information from the user profile, tweet content & community structure<br />Created Metadata<br /> After processing information in the user profile, content and/or network structure<br />57<br />
    • 67. An Example<br />Length: 109 charactersGeneral topic: Egypt protest <br />This poor {sentiment_expression: {target: “Lara Logan”, polarity: “negative”}} woman! RT @THRCBS News‘ {entity:{type=“News Agency”}} Lara Logan {entity:{type=“Person”}} Released FromHospital {entity:{type=“Hospital”}} After Egypt {entity:{type=“Country”} Assault {topic} http://bit.ly/dKWTY0 {external_URL}<br />58<br />
    • 68. Why Semantic Web is a Standard for Social Metadata?<br />Rich Snippet, open graph: RDFa - Semantic Web based social data standards<br />Relationships/connections play central role (not just hyperlinks as in Web data)– so relationship as first class object is important<br />Semantic Web technologies and standards provide better techniques to capture and represent metadata, relationships<br />59<br />
    • 69. Semantic Web in One Slide<br />Representing Semantic Web Data<br />RDF: relationships as first class object <subject, predicate,object><br />Representing Knowledge  and Agreements<br />nomenclature, taxonomy, folksonomy, ontology: OWL<br />Annotation: RDFa, Xlink, model reference<br />Web of Data: Linked Open Data <br />Querying: SPARQL <br />Rules: SWRL, RIF <br />60<br />
    • 70. How to Save and Use Metadata?<br />Store metadata as data and use standard database technique<br /> Use filtering and clustering, summarization, statistics - implicit semantics<br />61<br /><ul><li> Use explicit semantics and Semantic Web standardards and technologies
    • 71. Semantics = meaning
    • 72. Richer representation, support for relationships, context
    • 73. Supports use of background knowledge
    • 74. Better integration, powerful analysis
    • 75. Use of RDF data stores/LOD 
    • 76. Semantics- the implicit, the formal and the powerful
    • 77. Social metadata on the Web [H. Dacquin]</li></li></ul><li>Metadata Creation using Content Analysis<br />Building on foundations of <br />Statistical Natural Language Processing<br />Information Extraction<br />Semantic Web/ Knowledge Representation<br />62<br />
    • 78. Metadata Extraction from Informal Text<br />63<br />Meena Nagarajan,‘Understanding User-Generated Content on Social Media,’ Ph.D. Dissertation, Wright State University, 2010<br />
    • 79. 64<br />Characteristics of Text on Social Media<br />
    • 80. The Formality of Text<br />65<br />
    • 81. Content Analysis: Typical Sub-tasks<br /><ul><li>Sentiment Analysis
    • 82. What opinions are people conveying via the content?
    • 83. Author Profiling
    • 84. What can we infer about the author from the content he posts?
    • 85. Context (external to content) extraction
    • 86. URL extraction, analyzing external content</li></ul>Recognize key entities mentioned in content <br />Information Extraction (entity recognition, anaphora resolution, entity classification..)<br />Discovery of Semantic Associations between entities <br />Topic Classification, Aboutness of content <br />What is the content about? <br />Intention Analysis <br />Why did they share this content? <br />66<br />
    • 87. Research Efforts, Contributions in this space..<br />Examining usefulness of multiple context cues for text mining algorithms <br />Compensating for informal, highly variable language, lack of context<br />Using context cues: Document corpus, syntactic, structural cues, social medium, external domain knowledge…<br /> In this talk, highlighting sample metadata creation tasks: <br />NER<br />Key Phrase Extraction<br />Intention<br />Sentiment/Opinion Mining<br />67<br />
    • 88. Named Entity Recognition<br />I loved <movie> the hangover </movie>!<br />Key Phrase Extraction<br />68<br />Part 1: NER, Key Phrase Extraction<br />
    • 89. Multiple Context Cues Utilized for NER in Blogs and MySpace Forums<br />69<br />Meena Nagarajan,‘Understanding User-Generated Content on Social Media,’ Ph.D. Dissertation, Wright State University, 2010<br />
    • 90. 70<br />Multiple Context Cues Utilized for Keyphrase Extraction from Twitter, Facebook and MySpace<br />Meena Nagarajan,‘Understanding User-Generated Content on Social Media,’ Ph.D. Dissertation, Wright State University, 2010<br />
    • 91. Focus, Impact<br />We focus on techniques that exploit<br />content and context aspects on social media platforms<br />Our methods highlight a combination of top-down, bottom-up analysis for informal text<br />Statistical NLP, ML algorithms over large corpora (bottom-up)<br />Models and rich knowledge bases in a domain(top-down)<br />71<br />
    • 92. NAMED ENTITY RECOGNITION<br />72<br />
    • 93. Named Entity Recognition<br />“I loved your music Yesterday!” <br />Yesterday is an album<br />“It was THEHANGOVER of the year..lasted forever.. <br />The Hangover is not a movie<br />So I went to the movies..badchoice picking “GI Jane”worse now”<br />GI Jane is a movie<br />73<br />Task of NER : Identifying and classifying tokens<br />
    • 94. NER in prior work vs. NER for Informal Text<br />74<br />
    • 95. Cultural Named Entities<br />• NER focus in this work: Cultural Named Entities<br />Artifacts of Culture <br />Name of a books, music albums, films, video games, etc.<br />Common words in a language<br />The Lord of the Rings, Lips, Crash, Up, Wanted, Today, Twilight, Dark Knight…<br />75<br />
    • 96. What makes cultural entity extraction challenging..<br />Varied senses, several poorly documented<br />Star Trek: movies, TV series, media franchise.. and cuisines !!<br />Changing contexts with recent events<br />The Dark Knight is a movie, it is also a reference to Obamaand the health care policy<br />Comprehensive sense definitions, enumeration of contexts, labeled corpora for all senses .. Are Unrealistic expectationswhen building a NER system<br /> NER Relaxing the closed-world sense assumptions<br />76<br />
    • 97. 77<br />NER in prior work <br />vs. NER for Informal Text<br />
    • 98. A Spot and Disambiguate Paradigm<br />NER is generally a sequential prediction problem<br />NER system that achieves 90.8 F1 score on the CoNLL-2003 NER shared task (PER, LOC, ORGN entities)<br />78<br /><ul><li>Focus of approach: Spot and Disambiguate Paradigm
    • 99. Starting off with a dictionary or list of entities we want to spot
    • 100. Spot, then disambiguate in context (natural language, domain knowledge cues)
    • 101. Binary Classification
    • 102. Is this mention of “the hangover” in a sentence referring to a movie?</li></ul>CoNLL 2003 -- http://www.cnts.ua.ac.be/conll2003/ner/<br />
    • 103. 79<br />NER in prior work <br />vs. NER for Informal Text<br />
    • 104. Cultural NER - Two Flavors<br />80<br />
    • 105. (a) Multiple Senses in the Same Domain<br />81<br />
    • 106. Algorithm Preliminaries<br />Problem Definition<br />– Cultural Entity Identification : Music album, tracks<br /> e.g. Smile (Lilly Allen), Celebration (Madonna)<br />• Corpus: MySpace comments<br />– Context-poor utterances<br /> e.g. “Happy 25th Lilly, Alfieis funny”<br />82<br />• Goal: Semantic Annotation of music named entities (w.r.t MusicBrainz)<br />MusicBrainz Schema<br />
    • 107. Using a Knowledge Resource for NER is not straight-forward..<br />83<br />
    • 108. Approach Overview <br />Which ‘Merry Christmas’?; ‘So Good’is also a song!<br />Scoped Relationship graphs<br /> – Using context cues from the content, webpage title, url… <br />e.g. new Merry Christmas tune<br /> – Reduce potential entity spot size <br /> e.g. new albums/songs <br /> • Generate candidate entities<br /> • Spot and Disambiguate<br />84<br />
    • 109. Sample Real-world Constraints<br />Which ‘Merry Christmas’?; ‘So Good’is also a song!<br />Career Restrictions<br />- “release your third album already..”<br />Recent Album restrictions<br />- “I loved your new album..”<br />Artist age restrictions<br />-”happy 25thrihanna, loved alfie btw..”<br />etc.<br />85<br />
    • 110. 86<br />
    • 111. Scoping via Real-world Restrictions<br />87<br />
    • 112. Scoped Entity Lists<br />User comments are on MySpace artist pages<br /> – Contextual Restriction: Artist name<br /> – Assumption: no other artist/work mention<br />Naive spotter has advantage of spotting all possible mentions (modulo spelling errors)<br /> – Generates several false positives<br /> “this is bad news, ill miss you MJ”<br />88<br />
    • 113. But there are also non-music mentions<br />Challenge 1: Several senses in the same domain<br />Scoping relationship graphs narrows possible senses<br />Solves the named entity identification problem partially<br />Challenge 2: Non-music mentions<br />Got your new album Smile. Loved it!<br />Keep your SMILE on!<br />89<br />
    • 114. Using Language Features to eliminate incorrect mentions..<br />Syntactic features<br />POS Tags, Typed dependencies..<br />Word-level features<br />Capitalization, Quotes<br />Domain-level features<br />90<br />
    • 115. Supervised Learners<br />91<br />
    • 116. Hand-labeling - Fairly Subjective<br />1800+ spots in MySpace user comments from artist pages <br />Manual annotations for a post: “Keep your <track>SMILE<track>on!”<br />valid album/track named entity (good spot)invalid named entity (bad spot)hard-to tell (inconclusive)<br />4-way annotator agreements – shows that agreeing on the accuracy of a spot is hard to do even for domain experts<br /> – Madonna 90% agreement<br /> – Rihanna 84% agreement<br /> – Lily Allen 53% agreement (many named entities of ambiguous nature and usage)<br />92<br />
    • 117. Combining a Dictionary Spotter + NLP Analytics<br />93<br />Daniel Gruhl, Meena Nagarajan, Jan Pieper, Christine Robson, AmitSheth,‘Context and Domain Knowledge Enhanced Entity Spotting in Informal Text,’ The 8th International Semantic Web Conference, 2009: 260-276 <br />
    • 118. Lessons Learned - NER on Social Media Text using a Knowledge Base<br />Intelligent pruning of a knowledge base goes a long way in improving precision<br />Two stage approach: chaining NL learners over results of domain model based spotters<br />Improves accuracy up to a further 50%<br />allows the more time-intensive NLP analytics to run on less than the full set of input data<br />94<br />
    • 119. 95<br />Music NER application : BBC SoundIndex (IBM Almaden)Pulse of the Online Music Populace <br />Daniel Gruhl, MeenakshiNagarajan, Jan Pieper, Christine Robson, Amit Sheth: ‘Multimodal Social Intelligence in a Real-Time Dashboard System,’ special issue of the VLDB Journal on "Data Management and Mining for Social Networks and Social Media", 2010 <br />Project: http://www.almaden.ibm.com/cs/projects/iis/sound/<br />
    • 120. The Vision<br />http://www.almaden.ibm.com/cs/projects/iis/sound/<br />96<br />
    • 121. 97<br />
    • 122. Several Insights<br />98<br />Trending popularity of artists<br />Trending topics in artist pages<br />Only 4% -ve sentiments, perhaps ignore the Sentiment Annotator on this data source?<br />Ignoring Spam can change ordering <br />of popular artists<br />
    • 123. Predictive Power of Data<br />Billboards Top 50 Singles chart during the week of Sept 22-28 ’07 vs. MySpace popularity charts.<br />User study indicated 2:1 and upto 7:1 (younger age groups) preference for MySpace list.<br />Challenging traditional polling methods!<br />99<br />
    • 124. KEY PHRASE EXTRACTION<br />100<br />
    • 125. Key Phrase Extraction - Example<br />Key phrases extracted from prominent discussions on Twitter around the 2009 Health Care Reform debate and 2008 Mumbai Terror Attack on one day<br />101<br />
    • 126. Key Phrase Extraction from Social Media Text<br />Different from Information Extraction<br />Key phrase extraction does not concern itself with classification into a type<br />Extracting vs. Assigning Key Phrases <br />Focus: Key Phrase Extraction<br />Prior work focus: extracting phrases that summarize a document -- a news article, a web page, a journal article, a book..<br />Focus here: summarize multiple documents (UGC) around same event/topic of interest<br />102<br />
    • 127. Key Phrase Extraction on Social Media Content has some differences<br /> 1. Need to preserve/isolate the social behind the social data in <br /> summarizing key phrases<br /> What is said in Egypt vs. the USA should be viewed in isolation<br />2. Need to Accounting for redundancy, variability, off-topic content<br />“Met up with mom for lunch, she looks lovely as ever, good genes .. Thanks Nike, I love my new Gladiators ..smooth as a feather. I burnt all the calories of Italian joy in one run.. if you are looking for good Italian food on Main, Bucais the place to go.”<br />103<br />
    • 128. Where is the Social and Cultural Logic in UGC ?<br />Thematic components<br />similar messages convey similar ideas<br />Space, time metadata<br />role of community and geography in communication<br />Poster attributes<br />age, gender, socio-economic status reflect similar perceptions<br />‘Social applies to data as well as metadata’<br />104<br />
    • 129. Features used in social Key Phrase extraction (common to prior efforts)<br />Focus: n-grams, spatio-temporal metadata (social components) <br />Syntactic Cues: In quotes, italics, bold; in document headers; phrases collocated with acronyms<br />Document and Structural Cues: Two word phrases, appearing in the beginning of a document, frequency, presence in multiple similar documents etc. <br />Linguistic Cues: Stemmed form of a phrase, phrases that are simple and compound nouns in sentences etc.<br />105<br />
    • 130. Key Phrase Extraction Overview<br /> “President Obama in trying to regain control of the health-care debate will likely shift his pitch in September”<br />1-grams: President, Obama, in, trying, to, regain, ...<br /> 2-grams: “President Obama”, “Obama in”, “in trying”, “trying to”... <br /> 3-grams: “President Obama in”, “Obama in trying”; “in trying to”...<br />106<br />
    • 131. A descriptor is an n-gram weighted by:<br />Thematic Importance<br /><ul><li>TFIDF, stop words, noun phrases
    • 132. Redundancy: statistically discriminatory in nature
    • 133. Variability: contextually important</li></ul>Spatial Importance (local vs. global popularity)<br />Temporal Importance (always popular vs. currently trending)<br />`<br />107<br />
    • 134. 108<br />TF-IDF vs. Spatio-temporal-thematic scores rank phrases differently<br />Foreign relations surfaces up<br />M. Nagarajan et al., Spatio-Temporal-Thematic Analysis of Citizen-Sensor Data - Challenges and Experiences, Tenth International Conference on Web Information Systems Engineering, Oct 5-7, 2009: 539-553<br />
    • 135. Next task : Eliminating Off-topic Content<br />Frequency based heuristics will not eliminate off-topic content that is ALSO POPULAR<br />109<br />Popular Key phrases “single”, “Jesus” are unrelated to Madonna’s music<br />M. Nagarajan et al., Monetizing User Activity on Social Networks - Challenges and Experiences, 2009 IEEE/WIC/ACM International Conference on Web Intelligence, Sep 15-18 2009: 92-99<br />
    • 136. Elimination off-topic content : Example<br />• “Yeah i know this a bit off topic but the other electronics forum is dead right now. im looking for a good camcorder, somethin not to large that can record in full HD only ones so far that ive seen are sonys”<br />• “CanonHV20.Great little cameras under $1000.”<br />Possible relevant phrases are: ['camcorder', 'canon hv20', 'little camera', 'hd', 'cameras', 'canon']<br />110<br />
    • 137. • Assume one or more seed words (from domain knowledge base) C1 -['camcorder']<br />• Extracted Key words / phrasesC2 -['electronics forum', 'hd', 'camcorder', 'somethin', 'ive', 'canon', 'little camera', 'canon hv20', 'cameras', 'offtopic']<br />• Gradually expand C1 by adding phrases from C2 that are strongly associated with C1<br />• Mutual Information based algorithm [WISE2009]<br />111<br />Eliminating off-topic content : Approach Overview<br />
    • 138. Key Phrases & Aboutness - Evaluations<br />Are the key phrases we extracted topical and good indicators of what the content is about?<br />If it is, it should act as an effective index/search phrase and return relevant content<br />Evaluation Application: Targeted Content Delivery<br />112<br />
    • 139. Targeted Content Delivery -Evaluations<br />We took 12K posts from MySpace and Facebook Electronics forums<br />Extracted Baseline phrases using Yahoo Term Extractor<br />Extracted phrases using the Key phrase extraction, elimination algorithm described earlier<br />Generated Targeted Content from Google AdSense<br />Asked users if the delivered content matched the posts<br />113<br />
    • 140. Targeted Content for all content vs. extracted key phrases<br />114<br />
    • 141. User Studies and Results<br />115<br />
    • 142. Social Key Phrase Extraction : Impact, Contributions<br />TFIDF + social contextual cues yield more useful phrases that preserve social perceptions<br />Corpus + seeds from a domain knowledge base eliminate off-topic phrases effectively<br />116<br />
    • 143. INTENTION MINING<br />117<br />
    • 144. Why do people share?<br />Outside of the psychological incentives, broadly, people share to Seek Information OR Share Information<br />If we understand the intent behind a post, we can build systems that respond to it better<br />Focus of our work: Understand intent to deliver targeted content<br />Use case: Online Content-Targeted Advertisements on Social Media Platforms<br />118<br />
    • 145. Circa 2009 -Content-based Ads<br />119<br />
    • 146. Today – Content-based Ads on Profiles<br />120<br />
    • 147. What is going on here..<br /><ul><li>Ads are targeted on profile interests, demographic data
    • 148. But Interests on profiles do not translate to purchase intents</li></ul> – Interests are often outdated..<br /> – Intents are rarely stated on a profile.. <br />• Some profile data does seem to work<br />– Example: New store openings, sales targeted at location information in a profile<br />121<br />
    • 149. But Monetizable Intents are Elsewhere, away from their profiles..<br />122<br />
    • 150. Showing clear intents on MySpace posts but no relevant ads..<br />123<br />
    • 151. Targeted Content-based Advertizing <br />–Non-trivial<br /> –Non-policed content<br /> •Brand image, Unfavorable sentiments<br /> –People are there to network<br /> •User attention to ads is not guaranteed<br /> –Informal, casual nature of content<br /> •People are sharing experiences and events<br /> –Main message overloaded with off <br /> topic content<br />I NEED HELP WITHSONY VEGAS PRO 8!! Ugh and ihave a video project due tomorrow for merrilllynch :(( all ineed to do is simple: Extract several scenes from a clip, insert captions, transitions and thatsit. really. omggicant figure out anything!! help!! and igot food poisoning from eggs. its not fun. Pleasssse, help? :(<br />1Learning from Multi-topic Web Documents for Contextual Advertisement, Zhang, Y., Surendran, A. C., Platt, J. C., and Narasimhan, M.,KDD 2008 <br />124<br />
    • 152. Focus: Discuss Methodology, Preliminary Results in… <br />• Identifying intents behind user posts on social networks<br />– Identify Content with monetization potential<br />• Identifying keywords for advertizing in user-generated content<br />– Considering interpersonal communication & off-topic chatter<br />125<br />M. Nagarajan et al., ‘Monetizing User Activity on Social Networks - Challenges and Experiences,’ 2009 IEEE/WIC/ACM International Conference on Web Intelligence, Sep 15-18 2009: 92-99<br />
    • 153. Investigations<br />User studies<br /> – Hard to compare activity based ads to s.o.t.a<br /> – Impressions to Clickthroughs<br /> – How well are we able to identify monetizable posts<br /> – How targeted are ads generated using our <br /> keywords vs. entire user generated content<br />126<br />M. Nagarajan et al., ‘Monetizing User Activity on Social Networks - Challenges and Experiences,’ 2009 IEEE/WIC/ACM International Conference on Web Intelligence, Sep 15-18 2009: 92-99<br />
    • 154. Identifying Intents on SM is different from that on the Web..<br />Scribe Intent not same as Web Search Intent1<br />People write sentences, not keywords or phrases<br />Presence of a keyword does not imply navigational / transactional intents<br />– ‘am thinking of getting X’ (transactional)<br /> – ‘I like my new X’ (information sharing)<br /> – ‘what do you think about X’ (information seeking)<br />Useful here would be to identify: Transactional and Information Seeking intents<br />1B. J. Jansen, D. L. Booth, and A. Spink, “Determining the informational, navigational, and transactional intent of web queries,”Inf. Process. Manage., vol. 44, no. 3, 2008.<br />127<br />
    • 155. Not Focusing on the entity but Action Patterns surrounding the entity<br />“where can I find a chottopspcam”<br /> – User post also has an entity, which is a plus but not the main target of intent identification..<br />Goal is to study<br />How questions are asked and <br />not topic words that indicate what the question is about<br />128<br />
    • 156. Conceptual Overview Bootstrapping to learn IS patterns<br />Take a set of user posts from SNSs<br />Not annotated for presence or absence of any intent<br />129<br />
    • 157. Bootstrapping to learn IS patterns<br />Generate a universal set of n- gram patterns; freq > f<br />S = set of all 4-grams; freq > 3<br />130<br />
    • 158. Bootstrapping to learn IS patterns<br />Generate set of candidate patterns from seed words <br />(why,when,where,how,what)<br />Sc= all 4-grams in S that extract seed words<br />131<br />
    • 159. Bootstrapping to learn IS patterns<br />User picks 10 seed patterns from Sc<br />Sis= ‘does anyone know how’, ‘where do I find’, ‘someone <br />tell me where’… <br />132<br />
    • 160. Bootstrapping to learn IS patterns<br /> Gradually expand Sis by adding <br /> Information Seeking patterns from Sc<br />133<br />
    • 161. Bootstrapping to learn IS patterns<br />For every pis in Sis generate set of filler patterns<br />134<br />
    • 162. Bootstrapping to learn IS patterns<br />‘.* anyone know how’‘ does .* know how’<br /> ‘does anyone .* how’ ‘does anyone know .*’<br />Look for patterns in Sc<br />Functional compatibility of filler<br /><ul><li>words used in similar semantic contexts</li></ul>Empirical support for filler<br />135<br />
    • 163. Expanding the Pattern Pool<br />Functional properties / communicative functions of words<br />From a subset of LIWC1<br /> – cognitive mechanical (e.g., if, whether, wondering, find) <br /> • ‘I am thinking about getting X’ <br /> – adverbs(e.g., how, somehow, where) <br /> –(e.g., someone, anybody, whichever)<br /> • ‘Someone tell me where can I find X’ <br />1Linguistic Inquiry Word Count, LIWC, http://liwc.net<br />136<br />
    • 164. Example - Acquiring New Intent Patterns..<br />• ‘does * know how’<br />– ‘does someone know how’<br />• Functional Compatibility -Impersonal pronouns<br />• Empirical Support –1/3<br />– ‘does somebody know how’<br />• Functional Compatibility -Impersonal pronouns<br />• Empirical Support –0<br />• Pattern Retained<br />– ‘does john know how’<br />• Pattern discarded<br />Sc= {‘does anyone know how’, ‘where do I find’, ‘someone tell me where’}<br />• pis= `does anyone know how’<br />137<br />
    • 165. Finer Details of the Approach are in the paper..<br />Iterative algorithm, single-word substitutions, functional usage and empirical support conservatively expand the intent-seeking pool of patterns..<br />Infusing new patterns and seed words<br />Stopping conditions<br />138<br />M. Nagarajan et al., Monetizing User Activity on Social Networks - Challenges and Experiences, 2009 IEEE/WIC/ACM International Conference on Web Intelligence, Sep 15-18 2009: 92-99<br />
    • 166. Sample Extracted Patterns<br />139<br />
    • 167. Identifying Monetizable Posts<br />Information Seeking patterns just described are generated offline<br />Finding the Information seeking intent score of a post<br />– Extract and compare patterns in posts with extracted information seeking patterns<br /><ul><li>Transactional intent score of a post
    • 168. Using LIWC ‘Money’ dictionary : 173 words and word forms indicative of transactions, e.g., trade, deal, buy, sell, worth, price etc.</li></ul>140<br />
    • 169. Benchmarking with FB Marketplace<br />Training corpus<br />8000 user posts from MySpace Computers, Electronics, Gadgets forum<br /><ul><li>Information SeekingIntent patterns Extraction
    • 170. 309 unique new patterns, 263 unambiguous</li></ul>• Testing patterns for recall using ‘To buy’ Facebook Marketplace where all posts are information seeking<br />– extracted patterns average 81 % recall<br />141<br />
    • 171. Next task: Identifying Keywords for Advertizing<br />Identifying keywords in monetizable posts<br />– Plethora of work in this space<br />Off-topic noise removal is our focus<br />I NEED HELP WITHSONY VEGAS PRO 8!! Ugh and ihave a video project due tomorrow for merrilllynch :(( all ineed to do is simple: Extract several scenes from a clip, insert captions, transitions and thatsit. really. omggicant figure out anything!! help!! and igot food poisoning from eggs. its not fun. Pleasssse, help? :(<br />142<br />
    • 172. Conceptual Overview (also see slides in Key Phrase elimination section) <br /><ul><li>Topical hints</li></ul> – C1 -['camcorder']<br /><ul><li>Keywords in post</li></ul> – C2 -['electronics forum', 'hd', 'camcorder', 'somethin', 'ive', 'canon', 'little camera', 'canon hv20', 'cameras', 'offtopic']<br /><ul><li>Move strongly related keywords from C2 to C1 one-by-one</li></ul> – Relatedness determined using information gain<br /> – Using the Web as a corpus, domain independent<br />143<br />
    • 173. Example: Off-topic Chatter Elimination<br />• C1 -['camcorder']<br />• C2 -['electronics forum', 'hd', 'camcorder', 'somethin', 'ive', 'canon', 'little camera', 'canon hv20', 'cameras', 'offtopic']<br />• Informative words<br /> ['camcorder', 'canon hv20', 'little camera', 'hd', 'cameras', 'canon']<br />144<br />
    • 174. Evaluations- User Study<br />Keywords from 60 monetizable user posts<br /> – Monetizable intent, at least 3 keywords in content<br /> – 45 MySpace Forums, 15 Facebook Marketplace, 30 graduate students<br /> – 10 sets of 6 posts each<br /> – Each set evaluated by 3 randomly selected users<br />• Monetizable intents?<br /> – All 60 posts voted as unambiguously information seeking in intent<br />145<br />
    • 175. Effectiveness of using topical keywords<br />• Google AdSenseads for user post vs. extracted topical keywords<br />146<br />
    • 176. Instructions –User Study<br />147<br />
    • 177. Result -2X Relevant Impressions<br />Users picked ads relevant to the post<br /> – At least 50% inter-evaluator agreement<br />For the 60 posts<br />– Total of 144 ad impressions<br /> – 17% of ads picked as relevant<br />For the topical keywords<br /> – Total of 162 ad impressions<br />– 40% of ads picked as relevant<br />148<br />
    • 178. Evaluations: Profile Ads vs. Activity Ads<br />• Are ads generated from activity more interesting than those generated from user profiles?<br />Gather user’s profile information<br /> – Interests, hobbies, TV shows.. (non-demographic information)<br />• Ask them to submit a post (simulating their social media entry)<br /> – Looking to buy and why (induce off-topic content)<br />• Generate ads from profiles, from post (keywords)<br />149<br />
    • 179. Result - 8X more interest for non-profile ads..<br />• Using profile ads<br /> – Total of 56 ad impressions<br /> – 7% of ads generated interest<br />• Using authored posts<br /> – Total of 56 ad impressions<br /> – 43% of ads generated interest<br />• Using topical keywords from authored posts<br /> – Total of 59 ad impressions<br /> – 59% of ads generated interest<br />150<br />
    • 180. To note…<br /><ul><li>User studies small and preliminary, clearly suggest</li></ul> – Monetization potential in user activity<br /> – Improvement for Ad programs in terms of relevant impressions<br /><ul><li>Evaluations based on forum, marketplace</li></ul> – Verbose content<br /> – Status updates, notes, community and event memberships…<br /> – One size may not fit all<br />151<br />
    • 181. To note…<br />A world between relevant impressions and click throughs<br /> – Objectionable content, vocabulary impedance, Ad placement, network behavior<br /> – In a pipeline of other community efforts<br /><ul><li>No profile information taken into account</li></ul>– Cannot custom send information to Google AdSense<br />152<br />
    • 182. SENTIMENT / OPINION MINING<br />153<br />
    • 183. Content Analysis: Sentiment Analysis/Opinion Mining<br />Two main types of information we can learn from user-generated content: fact vs. opinion<br />Much of social media text (e.g., blogs, Twitter, Facebook) is a mix of facts and opinions.   <br />For example," Latest news: Mobile web services not working in #Bahrain and Internet is extremely slow #feb14{fact}... looks like they "learned" from #Egypt {opinion}"<br />
    • 184. Sentiment Analysis: Motivation<br />Why do people oppose health care reform?<br />What customers complain about?<br />Which movie should I see?<br />155<br />Image: http://bit.ly/eZtKBF<br />
    • 185. Sentiment Analysis: Tasks<br />Example: <br /> “How awful that many #Egypt ian artifacts are in danger of being Destroyed. What ZahiHawassmust be thinking#jan25”<br />Classification:<br />Overall sentiment polarity [Pang et al. 2002], [Turney 2002], etc.<br />the overall polarity is positive, neutral or negative (on the document/sentence/word level)<br />For the example: overall polarity is negative<br />Target-specific sentiment polarity [Yi et al. 2003], [Hu et al. 2004], etc. <br />The polarity toward the given target is positive, neutral or negative <br />For the Example: polarity is "negative“ for the target "egyptian artifacts“; polarity is "neutral“for target "ZahiHawass"<br />156<br />
    • 186. Sentiment Analysis: Tasks<br />Example: <br /> “How awful that many #Egypt ian artifacts are in danger of being Destroyed. What ZahiHawassmust be thinking #jan25”<br />Identification & Extraction: <br />opinion[Dave et al. 2003] etc.<br />opinion holder [Bethard et al. 2004] etc.<br />opinion target [Hu et al. 2004] etc.<br />For the example: <br /> opinion="awful", opinion holder="the author", target="egyptian artifacts are in danger”<br /> opinion="must be thinking", opinion holder="the author", target="ZahiHawass"<br />157<br />
    • 187. Sentiment Analysis: Classification<br />Supervised[Pang et al. 2002] etc.<br />Labeled training data: e.g., product review, movie review, etc.<br />Features: e.g., term-based, part-of-speech, syntactic relations, etc. <br />Learning strategies: e.g.,SVMs), Naive Bayes, ..<br />Unsupervised [Turney 2002] etc.<br />lexicon-based approach [Hu et al. 2004], [Ding et al. 2008] etc. <br />Using a sentiment lexicon of positive/negative sentiment words <br />Bootstrapping [Thelen et al. 2002] etc.<br />Iteratively trains and evaluates a classifier, starting from an unannotated corpus and a few predefined seed words, <br />The task of extracting the opinion/holder/target is similar to the traditional IE task. Key distinction- the relations between opinion and opinion target are considered important.<br />158<br />
    • 188. 159<br />Sentiment Analysis:<br />Identification & Extraction<br /><ul><li>Information extraction, utilizing the relations between opinion and opinion target
    • 189. Proximity[Hu et al. 2004] etc.
    • 190. extract the nearby adjectives modifying the target topic as opinion clues
    • 191. Syntactic dependency [Popescu et al. 2005] etc.
    • 192. employed language parser to compute the syntactic dependencies to extract the opinion clues with a given target topic
    • 193. Co-occurrence[Choi et al. 2009]etc.
    • 194. heuristics: the more frequently a candidate opinion target co-occurs with any opinion clues, the more likely it is the real opinion target
    • 195. Prepared patterns/rules [Kobayashi et al. 2004] etc.
    • 196. using a set of predefined extraction patterns/rules</li></li></ul><li>Sentiment Analysis: From Tweets to polls[O’Connor et al. 2010]<br />Findings:<br /><ul><li>A relatively simple sentiment detector based on tweets replicates consumer confidence and presidential job approval polls.
    • 197. Highlight the potential of text streams as a substitute and supplement for traditional polling.</li></ul>Connect public opinion measured from polls with sentiment measured from tweets. <br />Lexicon-based approachfor sentiment analysis of tweets<br />Within topic tweets, count messages containing positive and negative words defined by the sentiment lexicon<br />160<br />
    • 198. Sentiment Analysis: Predicting the Future With Social Media [Asur et al. 2010]<br />Use tweets to forecast box-office revenues for movies.<br />Traina language model classifier for sentiment classification of tweets.<br />Findings:<br />The prediction model using the rate at which tweets are created about a movie outperforms the market-based methods.<br />The sentiments present in tweets can be used to improve the prediction.<br />161<br />
    • 199. Sentiment Analysis: Target-specific Opinion Identification & Classification of Tweets-Unsupervised Approach [kno.e.sis ongoing work]<br />Simple lexicon-based method doesn't work well.<br />Target of “sexy” is “Helena”<br />Target of “terrific” is “reviews”<br />“free” is not opinionated in <br />movie domain. <br />Target of “loving” is “telling”<br />“well” in “as well” is not <br />opinionated<br />Observations:<br />The opinion clues may not be toward the given target (1,2,3,6)<br />The opinion clues are domain and context dependent (5,7)<br />Single words are not enough (4,7,8)<br />162<br />
    • 200. Domain and context-aware sentiment lexicon generation (here take the movie domain as example)<br />General subjective lexicon<br />Commonly used subjective lexicon + polar slangs learned from dictionary<br />Select candidate opinion clues from the domain-specific corpus based on the general lexicon<br />word + surrounding context<br />E.g., {“free”, “free movie”, “free movie streaming”... }, {“must”, “must see”, “a must see”, “must see movie”…} , {“well”, “as well”, “well done”… }<br />Identify the opinion clues and their polarity<br />Utilize information from multiple sources, including the corpus, domain knowledge (e.g., freebase, imdb), general lexicon, etc.<br />Bootstrapping + statistical model <br />E.g., <“must”, “must see”, positive>; <“well”, “well done”, positive> <br /> <br />Sentiment Analysis: Target-specific Opinion Identification & Classification of Tweets-Unsupervised Approach [kno.e.sis ongoing work]<br />163<br />
    • 201. 164<br />Sentiment Analysis: Target-specific Opinion Identification & Classification of Tweets-Unsupervised Approach [kno.e.sis ongoing work]<br /><ul><li>Target-specific opinion identification/extraction
    • 202. Predefined rules
    • 203. When generating the domain and context-aware sentiment lexicon, use a set of predefined rules to select toward-target candidate opinion clues
    • 204. Syntactic dependencies
    • 205. When using the generated lexicon to extract target-specific opinion, for each pair of <target, opinion clues> in one tweet, determine whether the opinion clues is toward the target based on their syntactic dependency.
    • 206. E.g., Lovedthe King's Speech. Funny, moving...Colin Firth is so amazing. I know, you already knew that. (“amazing” won’t be extracted since nsubj(amazing, Firth) )
    • 207. We also use predefined rules and proximity for complement </li></li></ul><li>PEOPLE ANALYSIS<br /><ul><li> Deriving People Metadata
    • 208. from Content Analysis
    • 209. from Network Analysis
    • 210. Merge of two approaches
    • 211. People Analysis showing use of Merger approach (Content+Network) and derived metadata
    • 212. Finding Influential Users
    • 213. Finding User Types & Affiliation
    • 214. Measuring Social Engagement </li></ul>165<br />
    • 215. People Analysis: Extracting People Metadata<br />166<br />
    • 216. People Analysis: Using Content to Derive People Metadata<br />Personality Signals<br />Extrovert, agreeable, open etc<br />Blogs, Style of Writing<br />Loose and periodic sentence, connotation etc.<br />Psychometric analysis of content<br />Knowledge, abilities, attitude etc.<br />Sample study: Gendered writing styles online [Ellison et al. 2006, Nagarajan et al. 2009, ICWSM etc.]<br />Self-expression tends towards attempting homophily in online dating profiles, given the tendency to 'imitate and impress' in courtship<br />167<br />Image: http://bit.ly/JZ6eF<br />Read: ‘How’ people write @Kno.e.sis<br />
    • 217. People Analysis: Using Network to Derive People Metadata<br />Interesting questions to ask:<br />Who are the most popular people* in the network<br />Who are the most influential people in the network<br />What are the types of people in the network<br />Who are the most active people in the communities<br />Who are the bridges between communities in the network, etc.<br />(*People may also refer to an organization)<br />168<br />Metadata from Network:<br /><ul><li>Not sufficient to answer the above questions for specific context and time, and hence merger approach (Content + Network) is better</li></ul> e.g., An Influential node in the network will be function of time and interest of his audience. <br />
    • 218. People Analysis: Influence<br />Adding Flavor of Context Analysis<br /><ul><li>Popularity NOT = Influence!
    • 219. For individuals to become influential they must not only obtain attention and thus be popular, but also overcome user passivity.  [Romero et al. 2010]
    • 220. Interest Similarity
    • 221. Homophily causing Reciprocity on Twitter [TwitterRank, Weng et al. 2010]
    • 222. Klout Score - True Reach, Amplification [http://klout.com]</li></ul>By Link Analysis Algorithms<br />Hits [Kleinberg 1999] & variants  <br />PageRank[Brin et al. 1998] & variants etc..<br />Links not sufficient!<br />Audience size doesn’t prove influence on twitter [Million Follower Fallacy,Cha et al. 2010]<br />169<br />Image: http://bit.ly/9pfTO4<br />
    • 223. People Analysis: User types & Affiliation<br />Blogger, Scientist, Journalist, Artist, Trustee, Company X in  Domain Y..<br />    - Multiple types and affiliations!<br />User interest mining<br />Key Phrase Extraction followed by semantic association on user bio, tweets, lists, favorite posts<br />Twitter Study [Banerjee et al. 2009]<br /><ul><li>Semantic analysis of profile description*
    • 224. Web Presence: Use of Web & Knowledge bases (Wikipedia, Blogs) to build context for user types
    • 225. Entity Spotting & Extraction, followed by Semantic Association and Similarity with user-type context</li></ul>170<br />Image: kahunainstitute.com<br />*Read Semantics driven Social Media Analysis@ Kno.e.sis<br />
    • 226. People Analysis: Social Engagement<br />171<br />Imagine a crisis scenario such as Haiti (2010) or Japan (2011) Earthquake<br /><ul><li> emergency teams are looking for ways to help the victims
    • 227. How effectively the community of people talking about this event online, can grow to reach potential donors and people in need of resources (food, water, first aids etc.)?
    • 228. What are the best possible ways to communicate between resource providers and people in need of resources?
    • 229. How teams can coordinate well between volunteers at a victim site, to managers in organizational structure, sitting in offices?</li></li></ul><li>People Analysis: Social Engagement<br />Can we find levels of user engagement* <br />Not just limiting to Passive vs. Active<br />Involvement in conversation, discussion & coordination <br />Mining Linguistics patterns & Platform specific handles in the conversations<br />Unsupervised learning approach [Ritter et al. 2010]<br />Frequency Distribution Analysis of user activity with added Domain Knowledge<br />posting, retweet, reply, mentions, lists etc. <br />172<br />Image: http://www.syscomminternational.com/<br />*More at Semantics driven Social Media Analysis@ Kno.e.sis<br />
    • 230. NETWORK ANALYSIS<br />- Deriving Network Metadata<br /> Interesting questions<br /> Network Analysis – Methods<br /> Models<br /> Metrics<br /> Network Analysis – Algorithms<br /> Graph Partitioning, Traversal<br /> Community Discovery, Evolution<br /> Social Network Analysis<br /> Diffusion <br />Homophily<br /> Study of 3-D Dynamics (People-Content-Network)<br />- Analysis & Visualization tools<br />173<br />
    • 231. Network Analysis <br />“To Discover How A, Who is in Touch with B and C, <br /> Is Affected by the Relation Between B & C”    <br /> -John Barnes<br />Interesting questions to ask:<br />How communitiesform around topics- growth & evolution<br />What are the effectsof presence of influential participants in the communities<br />What are the effectsof content nature (or sentiment, opinions) flowing in network on the community life<br />What is the community structure: degree of separation and sub-communities<br />174<br />Foundation of network: <br /><ul><li>Nodes
    • 232. Connections/Relationships</li></ul>Image: http://www.onasurveys.com/<br />
    • 233. Network Analysis: Methods<br />Network Modeling Approaches <br />Random graph model (Erdos-Renyi model)<br />start with n vertices and add edges between them at random<br />Small-world model<br />most nodes are not neighbors of one another, but they can be reached from every other by a small number of hops or steps (Small World Phenomenon)<br />Scale-free model <br />degree distribution follows a power law, i.e., frequency of degree varies as a power of its size<br />175<br />Image: http://www.kudosdynamics.com/<br />Important Literature:     <br />[Wasserman et al. 1992, Watts et al. 1998, Albert et al. 2002, Newman et al. 2006, Marin et al. 2010, Easley et al. 2010]<br />
    • 234. Network Analysis: Methods<br />176<br />Network Structure metrics<br />Centrality, Connected Component, Avg. Degree, Clustering Coefficient, Avg. Path Length, Bridge, Cohesion, Prestige, Reciprocity etc.<br />Social Network Analysis methods<br /><ul><li> Centrality (Degree, Eigenvector, Betweenness, Closeness)
    • 235. Clusters (Cliques and extensions, Communities)</li></ul>Image: http://www.kudosdynamics.com/<br />Important Literature:     <br />[Wasserman et al. 1992, Watts et al. 1998, Albert et al. 2002, Newman et al. 2006, Marin et al. 2010, Easley et al. 2010]<br />
    • 236. Network Analysis: Algorithms <br />Graph Partitioning & Traversal<br />Goal: Best time-complexity & reachability<br />Generally follows Greedy paths<br /> e.g., K-way multilevel Partitioning,<br />Bron-Kerbosch, K-plex, K-core or N-cliques, DFS, BFS, MST<br />Community Discovery, growth, evolution<br />Based on relationship types (e.g., signed network), geography/location based, interest based etc.<br />Generally follow cluster analysis<br /> e.g., Hierarchical clustering algorithms – Top-down, bottom-up<br /> Further Reading:<br />Modularity Maximization [Newman et al. 2006]<br />Algorithms comparison survey [Balakrishnan et al. 2006]<br />Online Communities [Preece 2001]<br />177<br />"We dream in Graph and We analyze in Matrix”<br />- Barry Wellman, INSNA <br />
    • 237. Social Network Analysis: Diffusion & Homophily<br />Social Network Analysis (Interested in information flow)<br /> Can we predict user actions? <br /> Understanding dynamics is challenging!<br />Why to study Diffusion<br />Maximizing Spread (Opinion, Innovation, Recommendation)<br />Outbreak Detection (e.g., disease)<br />Diffusion Behavior<br />Power Law distribution[Leskovec et al. 2007]<br />Factors impacting Diffusion<br />User Homophily – similar behavior tendency [McPherson et al. 2001]<br />Sampling strategy [Choudhury et al. 2010], content nature[Nagarajan et al. 2010]etc.<br />178<br />Image: http://bit.ly/fGkIBK<br />
    • 238. Study of 3-D Dynamics- People, Content, Network<br />Intra Community Activity and connectivity<br />How well connected are individual nodes (People)<br />What keeps them strongly connected over time (Relationship types - Knowledge of Content) <br />179<br />Will the two communities coordinate well during an event- crisis or disaster?<br /> - Interplay between all three dimensions – P, C, N<br /><ul><li>Inter-Community Connectivity
    • 239. Any bridges to connect to the other community? (People)
    • 240. Any Similarity in actions with the other community (Can Content help?)</li></ul>Image: http://themelis-cuiper.com<br />
    • 241. Study of 3-D Dynamics-People, Content, Network<br />Metadata a powerful tool to explore this dynamics*<br />Studies in this direction<br />A Qualitative Examination of Topical Tweet and Retweet Practices [Nagarajan et al. 2010] <br />How content dictates the network flow<br />User-Community Engagement by Multi-faceted Features: A Case Study on Twitter [Purohit et al. 2011] [TO BE PRESENTED TOMORROW IN SoME'11]<br />What factors impact user engagement in topic discussion<br />180<br />*Read People-Content-Network Analysis @ Kno.esis<br />
    • 242. Graphs showing sparse (A) and dense (B) RT networks and their corresponding follower graphs for 'call for action' and 'information sharing' type of tweets<br />M. Nagarajan, H. Purohit, and A. Sheth,  ’A Qualitative Examination of Topical Tweet and Retweet Practices,’ 4th Int'l AAAI Conference on Weblogs and Social Media, ICWSM 2010<br />181<br />
    • 243. Analysis & Visualization Tools<br />Network WorkBench (NWB)<br />Truthy <br />Graph-tool<br />Orange<br />Pajek<br />Tulip<br /> …. Many tools!!<br />Resource:<br />http://en.wikipedia.org/wiki/Social_network_analysis_software<br />182<br />Image:http://truthy.indiana.edu/<br />
    • 244. Citizen Sensing<br />Overview, Social Signals, Enablers<br />Role of Social Media<br />Activism, Journalism, Business Intelligence, Global Development<br />Development-Centric Platforms<br />Beginnings, Architectures and Possibilities<br />Systematic Study of Social Media<br />Spatio-Temporal-Thematic + People-Content-Network Analysis<br />Trustworthiness in Social Media<br />Mobile Social Computing<br />Citizen Sensing @ Real-time<br />Research Application: Twitris<br />Conclusion & Future Work<br />183<br />Outline<br />
    • 245. Trustworthiness in Social Media<br />Why?<br /><ul><li>Social media used for critical tasks
    • 246. In Disaster scenarios (e.g. Haiti earthquake, Gulf oil spill)
    • 247. For Political revolution (e.g. Egypt political crisis)
    • 248. In Political and Social policies (e.g. health care reforms) </li></ul>What?<br /><ul><li>Two step process for trustworthiness
    • 249. Remove off-topic content
    • 250. (How?) Detect spam and misleading content.
    • 251. Assess data quality of on-topic content
    • 252. (How?) Trust models to assess trustworthiness of content.</li></ul>184<br />
    • 253. Spam in Social Networks<br /><ul><li>Spamming is a major issue on social media
    • 254. Email spam drops while social media spam surges2
    • 255. e.g. spam on twitter
    • 256. 2% of [100 million]1 spam tweets per day!</li></ul>Graph depicting % of spam tweets <br />per day on twitter against time. <br />Image: http://blog.twitter.com/2010/03/state-of-twitter-spam.html<br />1http://techcrunch.com/2010/09/14/twitter-seeing-90-million-tweets-per-day/<br />1http://blog.twitter.com/2011/03/numbers.html<br />2http://www.allspammedup.com/2011/03/email-spam-drops-as-social-media-spam-surges/<br />185<br />
    • 257. Spam in Social Networks<br />Spamming on twitter:<br />Gaining Popularity<br />Use of popular topic related keywords (e.g. hashtags of trending topics) to propagate off topic content.<br />Launching malicious attacks<br />Phishing attacks, virus, malware etc.<br />Misleading the masses<br />Propagating false information [Mustafaraj et al. 2010]<br />Astroturf campaigns during political elections1.<br />1http://truthy.indiana.edu/<br />186<br />
    • 258. Spam in Social Networks<br />“By incorporating the hashtag #Cairo, which was being used to share updates about <br />the political protests taking place in Egypt, it appeared to many that he was using the <br />widely followed hashtag to selfishly promote the company’s clothing.” <br />- by Elisabeth Giammona1<br /><ul><li>Trending topic “abuse”</li></ul> Egypt Protests<br />Image: http://lat.ms/gHHfPg<br />Will topical consistency between the tweet and the URL help detect this?<br />Off-topic; Spam? Intelligent marketing?<br />1http://text100.com/hypertext/2011/02/what-not-to-do-with-hashtags-the-kenneth-cole-debacle/<br />187<br />
    • 259. Spam in Social Networks<br />Spam Detection<br />Content-based features<br />Content Size and repetition, URL usage, spam words [Benevenuto et al. 2010].<br />Metadata-based features<br />Account information, behavior [Ratkiewicz et al. 2010].<br />Network-based features <br />Provenance. (e.g. content from a reliable source)<br />Community voting. (e.g. RT in case of twitter).<br />Author status in the network (e.g. number of followers, friends, PageRank[Brin et al. 1998], influence [Weng et al. 2010]). <br />Removal of spam doesn’t guarantee trustworthiness <br />188<br />
    • 260. Trust in Social Networks<br />Reputation, Policy, Evidence, and Provenance used to derive trustworthiness [Artz et al. 2007].<br />Illustrative examples of online cues used for trust assessment<br />Wikipedia: article size, number of references, author, edit history, age of the article, edit frequency, etc [Dondi et al. 2006, Moturu et al. 2009].<br />Product Reviews: number of helpful, very helpful ratings, author expertise, sentiments in comments received for a review, etc [Liu et al. 2008].<br />189<br />
    • 261. Trust in Social Networks<br />Gleaning primitive (edge) trust <br />Trust value between two nodes is quantified using numbers. E.g., [0,1] or [-1,1] or partial ordering [Thirunarayan et al. 2009, Jøsang et al. 2002, Ganeriwal et al. 2008].<br />Gleaning composite (path) trust<br /> Propagation via chaining and aggregation (transitivity) [Golbeck et al. 2006, Sun et al. 2006]<br />Some popular algorithms for trust computation <br />Eigentrust[Kamvar et al. 2003], Spreading Activation [Ziegler et al. 2004], SUNNY [Kuter et al. 2007], etc.<br />190<br />
    • 262. Trust in Social Networks<br />Trust Ontology1<br />Captures semantics of trust.<br />Enables representation and reasoning with trust.<br />Semantics of Trust specifies, for a given trustor and trustee, the following features.<br />{type, value, scope, process}<br />trustor<br />trustee<br />1P. Anantharam, C. A. Henson, K. Thirunarayan, and A. P. Sheth, 'Trust Model for Semantic Sensor and Social Networks: A Preliminary Report', National <br />Aerospace & Electronics Conference (NAECON), Dayton Ohio, July 14-16th, 2010.<br />191<br />
    • 263. Trust Ontology<br /><ul><li>Trust Type1 - E.g., Referral Trust, Functional Trust and Non-Functional Trust.
    • 264. Referral Trust – Agent a1 trusts agent a2’s ability to recommend another agent.
    • 265. Functional Trust – Agent a1 trusts agent a2’s ability.
    • 266. Non-Functional Trust – Agent a1 distrusts agent a2’s ability.
    • 267. Trust Value - E.g., Star rating, numeric value or partial ordering.
    • 268. Trust Scope -E.g., recommendation for a Car Mechanic, Baby sitter or a movie.</li></ul>1K. Thirunarayan, Dharan K. Althuru, Cory A. Henson, and Amit P. Sheth, “A Local Qualitative Approach to Referral and Functional Trust,” <br />In: Proceedings of the The 4th Indian International Conference on Artificial Intelligence (IICAI-09), pp. 574-588, December 2009<br />192<br />
    • 269. Trust Ontology<br /><ul><li>Trust Process - Represents the process by which the Trust Value is computed and maintained.</li></ul>Reputation – based on past behavior.<br />Policy – based on explicitly stated constraints.<br />Evidence – based on seeking/verifying evidence. <br />Provenance – based on lineage information.<br />193<br />
    • 270. Trust Ontology<br />Anna’s car is in terrible shape<br />Dick is a certified mechanic<br />Bob has experience with cars<br />type: referral <br />process: reputation<br />scope: car mechanic<br />value: 8<br />Bob<br />type: functional <br />process: policy<br />scope: car mechanic<br />value: 10<br />ASE certified<br />type: non-functional<br />process: reputation<br />scope: car mechanic<br />value: 3<br />Ben<br />Dick<br />194<br />Anna<br />
    • 271. Citizen Sensing<br />Overview, Social Signals, Enablers<br />Role of Social Media<br />Activism, Journalism, Business Intelligence, Global Development<br />Development-Centric Platforms<br />Beginnings, Architectures and Possibilities<br />Systematic Study of Social Media<br />Spatio-Temporal-Thematic + People-Content-Network Analysis<br />Trustworthiness in Social Media<br />Mobile Social Computing<br />Citizen Sensing @ Real-time<br />Research Application: Twitris<br />Conclusion & Future Work<br />195<br />Outline<br />
    • 272. Mobile Social Computing<br />Instant Discovery: Geo-tagging and location-aware services, in combination with search, have made discovery a two-way street.<br />Compressed Expression: Mobile makes social networking even more compelling<br />Outsourced Memory: Cloud-based servers to store all of their mobile applications and databases<br />196<br /><ul><li>Automated Decisions: Smart apps helps to make faster decisions or even apps makes decisions for us
    • 273. Peer Power: Mobiles can create social movements based on peer influence</li></ul>http://www.technologyreview.com/business/26654/?a=f<br />
    • 274. Mobile Social Computing (Cont.)<br />Personalized Branding: advertising are rapidly becoming personalized based on individual's needs and preferences <br />Mobiles in social development becoming an integral part of development <br />Coordination in disaster situations<br />Health care delivery, especially in developing countries<br />Elections and other forms of political expression<br />Mobile payments will be widespread<br />Social Networking Accelerating Growth of Mobile<br />197<br />
    • 275. Integrating Social And Sensor Networks<br />Machine sensor observations are quantitative in nature, while human observations can be both qualitative and quantitative.<br />Benefits of combining observations from humans and machine sensors<br />Complementary evidence.<br />Corroborative evidence.<br />Applications of integrating heterogeneous sensor observations<br />Situation Awareness by using  human observations to interpret machine sensor observations.<br />Enhancing trustworthiness using corroborative evidence.<br />198<br />
    • 276. Citizen Sensing<br />Overview, Social Signals, Enablers<br />Role of Social Media<br />Activism, Journalism, Business Intelligence, Global Development<br />Development-Centric Platforms<br />Beginnings, Architectures and Possibilities<br />Systematic Study of Social Media<br />Spatio-Temporal-Thematic + People-Content-Network Analysis<br />Trustworthiness in Social Media<br />Mobile Social Computing<br />Citizen Sensing @ Real-time<br />Research Application: Twitris<br />Conclusion & Future Work<br />199<br />Outline<br />
    • 277. Citizen Sensing @ Real-time<br />200<br />
    • 278. Real-Time Motivation<br />People can’t wait for Information<br />500 years ago<br />Single life time<br />20 years ago<br />Next day or two<br />Television, News papers<br />Presently<br />Minutes are not considered <br /> fast enough<br />Digital media, Social media <br />201<br />Image: http://bit.ly/fg8EI3<br />
    • 279. Real-Time Social Media<br />Is Real-Time the future of Web?<br />Social Media for Real-Time Web<br />Disaster Management<br />Ushahidi (www.ushahidi.org)<br />Real-Time Markets<br />RealTimeMarkets (http://www.realtimemarkets.com/)<br />Brand Tracking<br />Twarql (http://wiki.knoesis.org/index.php/Twarql)<br />Movie reviews<br />Flicktweets (www.flicktweets.com)<br />202<br />
    • 280. Scenario<br />CNN - IBN<br />Feb 2011<br />Journalist<br />203<br />Scenario<br />Image: http://bit.ly/hlIutz<br />203<br />
    • 281. Challenges<br />Information Overload<br />Number of tweets that contained the words "Egypt," "Yemen," or "Tunisia" increased more than tenfold after January 23rd: there were 122,319 tweets between January 16 and 23 containing these terms, and 1.3 million tweets between January 24 and January 30.<br />Real Time<br />Can we extract social signals (through analysis) as data is generated?<br />http://blog.sysomos.com/2011/01/31/egyptian-crisis-twitte/<br />204<br />
    • 282. A Semantic Web Approach<br />Expressive description of Information need<br />Using SPARQL (Instead of traditional keyword search)<br /> Flexibility on the point of view<br />Ability to "slice and dice" the data in several dimensions: thematic, spatial, temporal, sentiment etc..<br />Streaming data with Background Knowledge<br />Enables automatic evolution and serendipity<br />Scalable Real-Time delivery <br />Using sparqlPuSH (SFSW'10)<br />205<br />
    • 283. Concept Feed<br />206<br />
    • 284. Architecture<br />Linked Open Social Signals @ Kno.e.sis<br />P. Mendes, A. Passant, P. Kapanipathi, and A. Sheth ‘Linked Open Social Signals,’ WI2010 IEEE/WIC/ACM International Conference on Web Intelligence (WI-10), Toronto, Canada, Aug. 31 to Sep. 3, 2010.<br />207<br />
    • 285. Social Sensor Server<br />208<br />
    • 286. Metadata Extractions  (Social Sensor Server)<br />Named Entity Recognition<br />2 Million Entities from DBPedia<br />Load as Trie for efficiency<br />N-grams matched<br />Example: Obama, Barack Obama<br />URL, HashTag Extraction<br />Regex extraction<br />Resolution<br />URL Resolution: Follows http redirects for resolution<br />HashTag Resolution: Tagdef, Tagal,WTHashTag.com<br />209<br /><ul><li>Other Metadata provided by Twitter
    • 287. User profile: User Name, Location, Time etc..
    • 288. Tweet: RT, reply etc..</li></li></ul><li>Structured Data(Social Sensor Server)<br />RDF Annotation<br />Common RDF/OWL Vocabularies<br />FOAF - (foaf-project.org) Friend of a Friend<br />SIOC - (sioc-project.org) Semantically Interlinked Online Communities<br />OPO - (online-presence.net) Online Presence Ontology<br />MOAT - (moat-project.org) — Meaning Of A Tag<br />210<br /><ul><li>A snippet of the annotation</li></ul><http://twitter.com/ bob/statuses/123456789><br />  rdf:type   sioct:MicroblogPost ;<br />  sioc:content  ”Fingers crossed for the upcoming #hcrvote” <br />  sioc:hascreator   <http://twitter.com/bob> ;<br />  foaf:maker    <http://example.org/bob> ;<br />  moat:taggedWith   dbpedia:Healthcare_reform .<br /><http://twitter.com/bob> geonames:locatedIn<br />Dbpedia:Ohio .<br />
    • 289. Semantic Publisher<br />211<br />
    • 290. Semantic Publisher<br />Virtuoso to filter triples<br />Queries formulated by the users are stored<br />SPARQL protocol over the HTTP to access rdf from the store<br />Combine data from tweet with the background knowledge in the rdf store <br />212<br />
    • 291. Application Server & Distribution Hub<br />213<br />
    • 292. Application Server & Distribution Hub<br />Distribution Hub<br /> PUSH Model - Pubsubhubbub protocol<br /> Pushes the tweets to the Application Server<br />Application Server<br /> Delivers data to the Clients<br /> RSS Enable Concept feeds<br />214<br />
    • 293. Use Case 4: Competitors<br />Brand Tracking - Competitors<br />Background Knowledge (e.g. DBpedia)<br />IPhone<br />HPTabletPC<br />category:Wi-Fi<br />@anonymized<br />This Ipad is super cool. Is it better than<br />category:Touchscreen<br />?category<br />skos:subject<br />skos:subject<br />?competitor<br />skos:subject<br />moat:taggedWith<br />dbpedia:IPad<br />?tweet<br />215<br />
    • 294. Demonstration<br />Social Media & News<br />Topic - HealthCare<br />1242 Articles from Nytimes<br />Around 800,000 tweets<br />President Obama lays out plan for Health care reform in Speech to Joint Session of Congress (10th Sept Timeline.com)<br />Obama taking an active role in Health talks in pursuing his proposed overhaul of health care system. (13th Aug Nytimes)<br />216<br />216<br />
    • 295. Twarql on Linked Open Data<br />217<br />Image: http://bit.ly/aTOV1r<br />217<br />
    • 296. Twarql (Real-Time Event Following)<br />Motivation - Rapidly Evolving Events<br />Example : Egypt protests, Health care debate, Iran Election<br />Twarql (Real-Time Event Following)<br />
    • 297. Adapt to the changes of the event<br />219<br />Approach – Continuous Semantics<br />Continuous Semantics @ Kno.e.sis<br />Sheth, C. Thomas, and P. Mehra, Continuous Semantics to Analyze Real-Time Data, <br />IEEE Internet Computing, 14 (6), November-December 2010, pp. 84-89. <br />
    • 298. Twarql (Real-Time Event Following) Twarql (Real-Time Event Following)<br />Twarql (Real-Time Event Following)<br />Twarql<br />Delivers tweets in Real-Time <br />Leverages background knowledge<br />Constraint (Sparql – Query)<br />Doozer<br />Mines Wikipedia <br />Automatically created domain model<br />http://wiki.knoesis.org/index.php/Twarql<br />http://knoesis.wright.edu/research/ModelCreation/<br />
    • 299. Continuous Semantics(Cycle)<br />Filter tweets (Twarql)<br />Leverage automatically created domain models<br />Extract Event descriptors (Twarql)<br />From the filtered tweets.<br />Create Domain models (Doozer)<br />Use extracted event descriptors and Back to (1).<br />Continuous Semantics(Cycle)<br />
    • 300. Twarql(Real-Time Event Following)<br />Architecture<br />New addition<br />
    • 301. Citizen Sensing<br />Overview, Social Signals, Enablers<br />Role of Social Media<br />Activism, Journalism, Business Intelligence, Global Development<br />Development-Centric Platforms<br />Beginnings, Architectures and Possibilities<br />Systematic Study of Social Media<br />Spatio-Temporal-Thematic + People-Content-Network Analysis<br />Trustworthiness in Social Media<br />Mobile Social Computing<br />Citizen Sensing @ Real-time<br />Research Application: Twitris<br />Conclusion & Future Work<br />223<br />Outline<br />
    • 302. Research Application: Twitris<br />224<br />
    • 303. Twitris - Motivation<br />Image: http://bit.ly/etFezl<br />1. Information Overload<br />Multiple events around us<br />WHAT to be aware of<br />Multiple Storylines about same event!!<br />225<br />
    • 304. Twitris - Motivation<br />2. Evolution of Citizen Observation<br />with location and time <br />226<br />
    • 305. Twitris - Motivation <br />3. Semantics of Social perceptions<br />What is being said about an event (theme)<br />Where (spatial)<br />When (temporal )<br />Twitrislets you browse citizen reports using social perceptions as the fulcrum<br />227<br />
    • 306. Twitris: Semantic Social Web Mash-up<br />Facilitates understanding of multi-dimensional social perceptions over SMS, Tweets, multimedia Web content, electronic news media<br />228<br />228<br />
    • 307. Twitris: Architecture<br />229<br />MeenakshiNagarajan, KarthikGomadam, AmitSheth, AjithRanabahu, RaghavaMutharaju and AshutoshJadhav, ‘Spatio-Temporal-Thematic Analysis of Citizen-Sensor Data - Challenges and Experiences,’ Tenth International Conference on Web Information Systems Engineering, 539 - 553, Oct 5-7, 2009.<br />
    • 308. Twitris: Functional Overview<br />230<br />
    • 309. Twitris: Event Summarization 1<br />231<br />
    • 310. Twitris: Event Summarization 2 <br />Sentiment Analysis <br />using statistical and machine learning techniques <br />232<br />
    • 311. Twitris: Event Summarization 3<br />Entity-relationship graph <br />using semantically annotated DBpedia entities mentioned in the tweets <br />233<br />
    • 312. Twitris: Demo, Quick Show <br />http://twitris.knoesis.org/<br />Many other interesting efforts – Eg: VivekK. Singh, MingyanGao, and Ramesh Jain. 2010. From microblogs to social images: event analytics for situation assessment. In Proceedings of the international conference on Multimedia information retrieval (MIR '10). ACM, New York, NY, USA, 433-436.<br />234<br />
    • 313. Citizen Sensing<br />Overview, Social Signals, Enablers<br />Role of Social Media<br />Activism, Journalism, Business Intelligence, Global Development<br />Development-Centric Platforms<br />Beginnings, Architectures and Possibilities<br />Systematic Study of Social Media<br />Spatio-Temporal-Thematic + People-Content-Network Analysis<br />Trustworthiness in Social Media<br />Mobile Social Computing<br />Citizen Sensing @ Real-time<br />Research Application: Twitris<br />Conclusion & Future Work<br />235<br />Outline<br />
    • 314. Citizen Sensing<br />Role of Social Media<br />Development-Centric Platforms<br />Systematic Study of Social Media<br />Spatio-Temporal-Thematic + People-Content-Network Analysis<br />Trustworthiness in Social Media<br />Mobile Social Computing<br />Citizen Sensing @ Real-time<br />Research Application: Twitris<br />236<br />Conclusion & Future Work<br />
    • 315. Twitris: Coordination<br />Great role in military and NGO rescue operations during emergencies: Haiti and Chile Earthquakes<br />237<br />
    • 316. Twitris: Coordination<br />Coordinating needs and resources in disaster situation<br />Analyze SMS and Web reports from disaster location<br />Use domain models for efficient and timely coordination<br />238<br />Image: http://bit.ly/hcp4PG<br />238<br />
    • 317. Twitris: CommunityFormation<br />Homophily in society <br />Bond of common interest <br />TRUST factor<br />239<br />
    • 318. Do you have a sense of immense opportunity of analyzing citizen sensing for useful social signals?<br />Do you appreciate the broad range of issues and challenges? Did we present examples and a few insights into how to address some unique challenges?<br />Did spatio-temporal-thematic and people-content-network dimensions present reasonable way to organize cast number of relevant research challenges and techniques?<br />Do you have more material to follow up on topics of interest?<br />240<br />Conclusions<br />
    • 319. References<br />[Mustafaraj et al. 2010] E. Mustafaraj and P. Metaxas, ‘From Obscurity to Prominence in Minutes: Political Speech and Real-Time Search’, In: Proceedings of the WebSci10: Extending the Frontiers of Society On-Line (April 2010).<br />[Anantharam et al. 2010] P. Anantharam, C. A. Henson, K. Thirunarayan, and A. P. Sheth, 'Trust Model for Semantic Sensor and Social Networks: A Preliminary Report', National Aerospace & Electronics Conference (NAECON), Dayton Ohio, July 14-16th, 2010.<br />[Thirunarayan et al. 2009] K. Thirunarayan, D. K. Althuru, C. A. Henson, and A. P. Sheth, 'A Local Qualitative Approach to Referral and Functional Trust,' In: Proceedings of the The 4th Indian International Conference on Artificial Intelligence (IICAI-09), pp. 574-588, December 2009.<br />[Gruhl et al. 2010] D. Gruhl, M. Nagarajan, J. Pieper, C. Robson, A. Sheth, ‘Multimodal Social Intelligence in a Real-Time Dashboard System,’ in a special issue of the VLDB Journal on 'Data Management and Mining for Social Networks and Social Media, 2010.<br />241<br />
    • 320. References 2<br />[O’Connor et al. 2010] B.O’Connor, R.Balasubramanyan, B.R.Routledge, and N.A.Smith, ‘From Tweets to polls: Linking text sentiment to public opinion time series’, In International AAAI Conference on Weblogs and Social Media, Washington,D.C., 2010.<br />[Asur et al. 2010] S. Asur and B. A.Huberman, ‘Predicting the Future With Social Media’, 2010. http://arxiv.org/abs/1003.5699<br />[Sheth et al. 2009] A. Sheth, ‘Semantic Integration of Citizen Sensor Data and Multilevel Sensing: A comprehensive path towards event monitoring and situational awareness’, February 17, 2009.<br />[Brin et al. 1998] S. Brin and L. Page, ‘The anatomy of a large-scale hypertextual Web search engine,’ Computer Networks and ISDN Systems, Vol 30, 1-7, 1998.<br />242<br />
    • 321. References 3<br />[Sheth et al. 2010] A. Sheth, C. Thomas, and P. Mehra, ‘Continuous Semantics to Analyze Real-Time Data,’ IEEE Internet Computing, November-December 2010, pp. 80-85.<br />[Nagarajan et al. 2010] M. Nagarajan, H. Purohit, and A. Sheth,  ’A Qualitative Examination of Topical Tweet and Retweet Practices,’ 4th Int'l AAAI Conference on Weblogs and Social Media, ICWSM 2010.<br />[Romero et al. 2010] D. Romero, W. Galuba, S. Asur, and B. Huberman, ‘Influence and Passivity in Social Media,’ Arxiv preprint, arXiv:1008.1253, 2010.<br />[Leskovec et al. 2009] J. Leskovec, K. Lang, A. Dasgupta, and M. Mahoney. ‘Community structure in large networks: Natural cluster sizes and the absence of large well-defined clusters’, Internet Mathematics, 6(1):29{123, 2009.<br />[Cha et al. 2010] M. Cha, H. Haddadi, F. Benevenuto, and K. Gummadi, ‘Measuring user influence in twitter: The million follower fallacy,’ In ICWSM'04, 2010.<br />[Pang et al. 2008] B. Pang and L. Lee. ‘Opinion mining and sentiment analysis,’ Foundations and Trends in Information Retrieval, 2(1-2):1-135, 2008. <br />243<br />
    • 322. References 4<br />[Pang et al. 2002] B. Pang, L. Lee, and S. Vaithyanathan, ‘Thumbs up? Sentiment classification using machine learning techniques,’ In Proceedings of EMNLP, pages 79–86, 2002.<br />[Turney 2002] P. Turney, ‘Thumbs up or thumbs down? Semantic orientation applied to unsupervised classification of reviews,’ In Proceedings of ACL, pages 417–424, 2002.<br />[Thelen et al. 2002] M. Thelen and E. Riloff, ‘A bootstrapping method for learning semantic lexicons using extraction pattern contexts,’ In Proceedings of EMNLP, pages 214-221, 2002.<br />[Yi et al. 2003] J. Yi, T. Nasukawa, R. Bunescu, and W. Niblack, ‘Sentiment analyzer: Extracting sentiments about a given topic using natural language processing techniques,’ In Proceedings of ICDM, 2003.<br />[Nagarajan et al. 2009, WI] M. Nagarajan, K. Baid, A. Sheth, and S. Wang, ‘Monetizing User Activity on Social Networks - Challenges and Experiences,’ 2009 IEEE/WIC/ACM International Conference on Web Intelligence, Sep 15-18 : pp. 92-99, 2009<br />244<br />
    • 323. References 5<br />[Hu et al. 2004] M. Hu and B. Liu, ‘Mining and summarizing customer reviews,’ In Proceedings of the ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), pages 168–177, 2004.<br />[Ding et al. 2008] X. Ding, B. Liu, and P. S. Yu, ‘A holistic lexicon-based approach to opinion mining,’ In Proceedings of WSDM, 2008.<br />[Jijkoun et al. 2010] V. Jijkoun, M. d. Rijke, and W. Weerkamp, ‘Generating Focused Topic-specific Sentiment Lexicons,’ In Proceedings of ACL, 2010.<br />[Bethard et al. 2004] S. Bethard, H. Yu, A. Thornton, V. Hatzivassiloglou, and D. Jurafsky, ‘Automatic extraction of opinion propositions and their holders,’ In Proceedings of the AAAI Spring Symposium on Exploring Attitude and Affect in Text, 2004.<br />[Hu et al. 2004] M. Hu and B. Liu, ‘Mining opinion features in customer reviews,’ In Proceedings of AAAI, pages 755-760, 2004.<br />[Dave et al. 2003] K. Dave, S. Lawrence, and D. M.Pennock, ‘Mining the peanut gallery: Opinion extraction and semantic classification of product reviews,’ In Proceedings of WWW, pages 519–528, 2003.<br />245<br />
    • 324. References 6<br />[Popescu et al. 2005] A.-M. Popescu and O. Etzioni, ‘Extracting product features and opinions from reviews,’ In Proceedings of HLT/EMNLP, 2005.<br />[Choi et al. 2009] Y. Choi, Y. Kim, and S.-H. Myaeng, ‘Domain-specific sentiment analysis using contextual feature generation,’ In Proceeding of the International CIKM workshop on Topic-sentiment analysis for mass opinion (TSA), 2009.<br />[Kobayashi et al. 2004] N. Kobayashi, K. Inui, Y. Matsumoto, K. Tateishi, and T. Fukushima, ‘Collecting evaluative expressions for opinion extraction,’ In Proceedings of the International Joint Conference on Natural Language Processing (IJCNLP), 2004<br />[A. Sheth 2009] A. Sheth, 'Citizen Sensing,Social Signals, andEnriching Human Experience', IEEE Internet Computing, July/August 2009, pp. 80-85. <br />[Nagarajan et al. 2009, WISE] M. Nagarajan, K. Gomadam, A. Sheth, A. Ranabahu, R. Mutharaju, and A. Jadhav, ‘Spatio-Temporal-Thematic Analysis of Citizen-Sensor Data - Challenges and Experiences,’ Tenth International Conference on Web Information Systems Engineering, 539 - 553, Oct 5-7, 2009.<br />246<br />
    • 325. References 7<br />[Kleinberg 1999] J. Kleinberg, ‘Authoritative sources in a hyperlinked environment,’ Journal of the ACM 46 (5): 604 -632, 1999.<br />[Albert et al. 2002] R. Albert and A.L. Barabasi, ‘Statistical Mechanics of Complex Networks,’ Rev. Modem Physics,’ vol. 74, no. 1, pp. 47-97, 2002.<br />[Weng et al. 2010] J. Weng and E. Lim and J. Jiang, and Q. He, ‘TwitterRank: nding topic-sensitive influential twitterers,’ WSDM, 2010.<br />[Banerjee et al. 2009] N. Banerjee, D. Chakraborty, K. Dasgupta, S. Mittal, A. Joshi, S. Nagar, A. Rai, and S. Madan, ‘User interests in social media sites: an exploration with micro-blogs,’ CIKM '09.<br />[Ritter et al. 2010] A. Ritter, C. Cherry, and B. Dolan, ‘Unsupervised modeling of Twitter conversations,’ InHuman Language Technologies: ACL (HLT '10), 2010.<br />[Watts et al. 1998] D.J. Watts and S.H. Strogatz, ‘Collective dynamics of 'small-world' networks,’ Nature 393 (6684): 409–10, 1998<br />[Newman et al. 2006] M. E. J. Newman, D. J. Watts, ‘The structure and dynamics of network,’ Princeton University Press, 2006<br />[Wasserman et al. 1992] Wasserman and Faust, ‘Social Network Analysis’, 1992<br />247<br />
    • 326. References 8<br />[Easley et al. 2010] D. Easley, J. Kleinberg, ‘Networks, Crowds, and Markets: Reasoning About a Highly Connected World,’ Cambridge University Press, 2010.<br />[Marin et al. 2010] A. Marin and B. Wellman, ‘Handbook of Social Network,’ Analysis, 2010.<br />[Balakrishnan 2006] H. Balakrishnan, ‘Algorithms for Discovering Communities in Complex Networks. Ph.D. Dissertation. University of Central Florida,’ Orlando, FL, USA. Advisor(s) NarsinghDeo. 2006.<br />[Choudhury et al. 2010] M. D. Choudhury, Y-R. Lin, H. Sundaram, K. S. Candan, L. Xie, and A. Kelliher, ‘How Does the Sampling Strategy Impact the Discovery of Information Diffusion in Social Media?’, ICWSM 2010<br />[Leskovec et al. 2007] J. Leskovec, L. A. Adamic, and B. A. Huberman, ‘The dynamics of viral marketing,’ ACM Trans. Web 1, 1, Article 5, May 2007.<br />[Purohit et al. 2011] H. Purohit, Y. Ruan, A. Joshi, S. Parthasarathy and A. Sheth. Understanding User-Community Engagement by Multi-faceted Features: A Case Study on Twitter. To appear in SoME'11 (Workshop on Social Media Engagement, in conjunction with WWW 2011)<br />248<br />
    • 327. References 9<br />[Pablo et al. 2010] P. Mendes, A. Passant, P. Kapanipathi, and A. Sheth‘Linked Open Social Signals,’ WI2010 IEEE/WIC/ACM International Conference on Web Intelligence (WI-10), Toronto, Canada, Aug. 31 to Sep. 3, 2010.<br />[Pablo et al. 2010] P. Mendes, P. Kapanipathi, and A. Passant, ‘Twarql: Tapping into the Wisdom of the Crowd,’. Triplification Challenge 2010 at 6th International Conference on Semantic Systems (I-SEMANTICS), Graz, Austria, 1-3 September 2010. (Winner of Triplification Challenge 2010)<br />http://www.nascio.org/events/2009Midyear/documents/NASCIO-KeynoteNoveck.pdf<br />http://citizensensing.posterous.com<br />[Ellison et al. 2006] N. Ellison, R. Heino, and J. Gibbs. ‘Managing Impressions Online: Self-Presentation Processes in the Online Dating Environment’. Journal of Computer-Mediated Communication 11(2):415–441. 2006<br />249<br />
    • 328. References 10<br />[Nagarajan et al. 2009, ICWSM] M. Nagarajan, M. A. Hearst. ‘An Examination of Language Use in Online Dating Personals’, 3rd Int'l AAAI Conference on Weblogs and Social Media, 266-269, ICWSM 2009.<br />[McPherson et al. 2001] M. McPherson, L. Smith-Lovin and J. M Cook, ‘Birds of a Feather: Homophily in Social Networks’, Annual Review of Sociology, Vol. 27: 415 -444. 2001<br />[Preece 2000] J. Preece. ‘Online communities: Usability, Sociability, Theory and Methods’. Frontiers of Human-Centred Computing, Online Communities and Virtual Environments, 2001<br />[Kashyap et al. 1998] V. Kashyap and A. Sheth, 'Semantic Heterogeneity in Global Information Systems: The Role of Metadata, Context and Ontologies,’ in Cooperative Information Systems: Current Trends and Directions, M. Papazoglou and G. Schlageter (Eds.), Academic Press, 1998, pp. 139-178.<br /><ul><li>[Kuter et al. 2007] U. Kuter and J. Golbeck, ‘SUNNY: a new algorithm for trust inference in social networks using probabilistic confidence models,’ Proceedings of the 22nd national conference on Artificial intelligence, Vol. 2, pp. 1377--1382, 2007.</li></ul>250<br />
    • 329. References 10<br />[Dondi et al. 2006] P. Dondio, S. Barrett, S. Weber, and J. Seigneur, ‘Extracting Trust from Domain Analysis: A Case Study on the Wikipedia Project,’ In: Autonomic and Trusted Computing In Autonomic and Trusted Computing, Vol. 4158 (2006), pp. 362-373.<br />[Liu et al. 2008] H. Liu, E. P. Lim, H. W. Lauw, M. T. Le, A. Sun, J. Srivastava, Y. A. Kim, ‘Predicting trusts among users of online communities: an epinions case study,’ In: EC '08: Proceedings of the 9th ACM conference on Electronic com-merce (2008), pp. 310-319.<br />[Benevenuto et al. 2010] F. Benevenuto, G. Magno, T. Rodrigues, V. Almeida, ‘Detecting Spammers on Twitter,’ In Collaboration, Electronic messaging, Anti-Abuse and Spam Conference (CEAS) (July 2010).<br />[Ratkiewicz et al. 2010] J. Ratkiewicz, M. Conover, M. Meiss, B. Gonalves, S. Patil, A. Flammini, F. Menczer, ‘Detecting and Tracking the Spread of As-troturf Memes in Microblog Streams,’ CoRR, Vol. abs/1011.3768, 16th November 2010.<br />[Ziegler et al. 2004] C.-N. Ziegler and G. Lausen, ‘Spreading Activation Models for Trust Propagation,’ In Proceedings of the IEEE International Conference on e-Technology, e-Commerce, and e-Service, 2004.<br />251<br />
    • 330. References 10<br />[Moturu et al. 2009] S. T. Moturu and H. Liu, ‘Evaluating the Trustworthiness of Wikipedia Articles through Quality and Credibility’, The 5th International Symposium on Wikis and Open Collaboration, Florida, Oct 25-27, 2009.<br />[Jøsang et al. 2002] A. Jøsang and R. Ismail, ‘The Beta Reputation System,’ Proceedings of the 15th Bled Conference on Electronic Commerce, Bled, Slovenia, 17-19 June 2002.<br />[Ganeriwal et al. 2008] S. Ganeriwal, L. K. Balzano, and M. B. Srivastava, ‘Reputation-based frameworkfor high integrity sensor networks,’ ACM Trans. Sen. Netw., Volume 4, issue 3, 1-37,June, 2008.<br />[Sun et al. 2006] Y. L. Sun, W. Yu, Z. Han, and K.J.R. Liu, ‘Information theoretic framework of trust modeling and evaluation for ad hoc networks,’ Selected Areas in Communications, IEEE Journal on , vol.24, no.2, pp. 305- 317, Feb. 2006<br />[Golbeck et al. 2005] J. Golbeck and J. Hendler, 'Inferring Trust Relationships in Web-Based Social Networks,' ACM Transactions on Internet Technology, Vol. 7, 2005.<br />252<br />
    • 331. About us<br />MeenaNagarajan is a research staff scientist at the IBM Almaden Research Center. Meena completed her dissertation on "Understanding User-Generated Content on Social Media" in 2010 at Wright State University Center of Excellence on Knowledge-enabled Computing (Kno.e.sis). She has collaborated with IBM Research, Microsoft Research, Marti Hearst at University of Berkeley and HP Labs. She has extensively published and served on Program Committees for key conferences. She also has the rare distinction of being invited to give a keynote (http://knoesis.org/library/resource.php?id=731), chair a panel before completing her Ph.D., and was selected for the prestigious NSF CI Fellows award. Her research has played a key role in state of the art social media and analytics systems such as BBC SoundIndex.<br /> <br />Amit Sheth is an educator, researcher and entrepreneur. He is the LexisNexis Ohio Eminent Scholar at the Wright State University, Dayton OH. He directs Kno.e.sis - the Ohio Center of Excellence in Knowledge-enabled Computing (http://knoesis.org, http://bit.ly/coe-k) which works on topics in Semantic, Social, Sensor, and Services computing over the Web, with the goal of advancing from the information age to meaning age. Prof. Sheth is an IEEE fellow and is one of the most highly cited authors in Computer Science (h-index = 68, http://bit.ly/CS-h) and World Wide Web (http://bit.ly/mas-www). He is Editor-in-Chief of the International Journal of Semantic Web & Information Systems, joint- Editor-in-Chief of Distributed & Parallel Databases, series co-editor of two Springer book series and serves on several editorial boards. By licensing his funded university research, he has also founded and managed two successful companies. Several commercial products and many operationally deployed applications have resulted from his R&D. More about Amit: http://knoesis.org/amit<br /> <br />SelvamVelmurugan is a technology entrepreneur, social entrepreneur and a leader/evangelist in use of social technology for development. He served in technology and leadership positions at Amazon and led the establishment of Amazon's development center in India.  Then he founded eMoksha (emoksha.org), a non-partisan non-profit organization focused on enabling citizen awareness and engagement using internet, mobile technologies and Kiirti (kiirti.org), a platform to enable effective governance by promoting awareness and citizen engagement by engaging citizens through phone, SMS, email and the web. eMoksha and Kiirti projects in developing countries are being studied by organizations around the world (including Harvard and Columbia) to  understand the role and impact of technology in enabling transparency  and civic engagement. Selvam was selected as a Center for Internet and Society (India) Fellow and was invited to attend TEDIndia in 2009. More about Selvam: http://www.ted.com/profiles/view/id/306995<br />253<br />

    ×