Evolving social data mining and affective analysis methodologies, framework and applications - Web 2.0 facts and social data
Social associations and all kinds of graphs
Evolving social data mining
Emotion-aware social data analysis
Frameworks and Applications
A glimpse into what social media is all about and how the researchers in the world are using social media. Social media is not a mere hype and not a platform to leverage word-of-the-mouth practices as is the common perception of it in Pakistan: it is much more than that and this is what this talk presented.
The growth of social media over the last decade has revolutionized the way individuals interact and industries conduct business. Individuals produce data at an unprecedented rate by interacting, sharing, and consuming content through social media. Understanding and processing this new type of data to glean actionable patterns presents challenges and opportunities for interdisciplinary research, novel algorithms, and tool development. Social Media Mining integrates social media, social network analysis, and data mining to provide a convenient and coherent platform for students, practitioners, researchers, and project managers to understand the basics and potentials of social media mining. It introduces the unique problems arising from social media data and presents fundamental concepts, emerging issues, and effective algorithms for network analysis and data mining. Suitable for use in advanced undergraduate and beginning graduate courses as well as professional short courses, the text contains exercises of different degrees of difficulty that improve understanding and help apply concepts, principles, and methods in various scenarios of social media mining.
Details at: http://dmml.asu.edu/smm/
A glimpse into what social media is all about and how the researchers in the world are using social media. Social media is not a mere hype and not a platform to leverage word-of-the-mouth practices as is the common perception of it in Pakistan: it is much more than that and this is what this talk presented.
The growth of social media over the last decade has revolutionized the way individuals interact and industries conduct business. Individuals produce data at an unprecedented rate by interacting, sharing, and consuming content through social media. Understanding and processing this new type of data to glean actionable patterns presents challenges and opportunities for interdisciplinary research, novel algorithms, and tool development. Social Media Mining integrates social media, social network analysis, and data mining to provide a convenient and coherent platform for students, practitioners, researchers, and project managers to understand the basics and potentials of social media mining. It introduces the unique problems arising from social media data and presents fundamental concepts, emerging issues, and effective algorithms for network analysis and data mining. Suitable for use in advanced undergraduate and beginning graduate courses as well as professional short courses, the text contains exercises of different degrees of difficulty that improve understanding and help apply concepts, principles, and methods in various scenarios of social media mining.
Details at: http://dmml.asu.edu/smm/
Social Targeting: Understanding Social Media Data Mining & AnalysisInfini Graph
Â
Chase McMichael â CEO, InfiniGraph
Social Targeting: Understanding Social Media Data Mining & Analysis
With the advent of the social web, companies that arenât actively mining, analysing and using social media data are missing a huge commercial advantage. In this session Chase McMichael will explain how social targeting works, including technologies, techniques and opportunities. He will also highlight the privacy challenges facing the industry.
Fuzzy AndANN Based Mining Approach Testing For Social Network AnalysisIJERA Editor
Â
Fast and Appropriate Social Network Analysis (SNA) tools ,techniques, are required to collect and classify
opinion scores on social networksites , as a grouping on wrong opinion may create problems for a society or
country . Social Network Analysis (SNA) is popular means for researcher as the number of users and groups
increasing day by day on that social sites , and a large group may influence other.In this paper, we
recommendhybrid model of opinion recommendation systems, for single user and for collective community
respectively, formed on social liking and influence network theory. By collecting thedata of user social networks
and preferenceslike, we designed aimproved hybrid prototype to imitate the social influence by like and sharing
the information among groups.The significance of this paper to analyze the suitability of ANN and Fuzzy sets
method in a hybrid manner for social web sites classifications, First, we intend to use Artificial Neural
Network(ANN)techniques in social media data classification by using some contemporary methods different
than the conventional methods of statistics and data analysis, in next we want to propagate the fuzzy approach
as a way to overcome the uncertainity that is always present in social media analysis . We give a brief overview
of the main ideas and recent results of social networks analysis , and we point to relationships between the two
social network analysis and classification approaches .This researchsuggests a hybrid classification model build
on fuzzy and artificial neural network (HFANN). Information Gain and three popular social sites are used to
collect information depicting features that are then used to train and test the proposed methods . This neoteric
approach combines the advantages of ANN and Fuzzy sets in classification accuracy with utilizing social data
and knowledge base available in the hate lexicons.
Sampling of User Behavior Using Online Social NetworkEditor IJCATR
Â
The popularity of online networks provides an opportunity to study the characteristics of online social network graphs is important, both to improve current systems and to design new application of online social networks. Although personalized search has been proposed for many years and many personalization strategies have been investigated, it is still unclear whether personalization is consistently effective on different queries for different users, and under different search contexts. In this paper, we study performance of information collection in a dynamic social network. By analyzing the results, we reveal that personalized search has significant improvement over common web search.
The mixing time of thee sampling process strongly depends on the characteristics of the graph.
R. Zafarani, M. A. Abbasi, and H. Liu, Social Media Mining: An Introduction, Cambridge University Press, 2014.
Free book and slides at http://socialmediamining.info/
Avoiding Anonymous Users in Multiple Social Media Networks (SMN)paperpublications3
Â
Abstract: The main aim of this project is secure the user login and data sharing among the social networks like Gmail, Facebook and also find anonymous user using this networks. If the original user not available in the networks, but their friends or anonymous user knows their login details means possible to misuse their chats. In this project we have to overcome the anonymous user using the network without original user knowledge. Unauthorized user using the login to chat, share images or videos etc This is the problem to be overcome in this project .That means user first register their details with one secured question and answer. Because the anonymous user can delete their chat or data In this by using the secured questions we have to recover the unauthorized user chat history or sharing details with their IP address or MAC address. So in this project they have found out a way to prevent the anonymous users misuse the original user login details.
Introduction to the Responsible Use of Social Media Monitoring and SOCMINT ToolsMike Kujawski
Â
These are my slides from a custom tool-based demonstration workshop I was asked to do where I went over various free tools that can be used to obtain valuable public data.
Social Media Mining - Chapter 6 (Community Analysis)SocialMediaMining
Â
R. Zafarani, M. A. Abbasi, and H. Liu, Social Media Mining: An Introduction, Cambridge University Press, 2014.
Free book and slides at http://socialmediamining.info/
Predicting Social Interactions from Different Sources of Location-based Knowl...Michael Steurer
Â
Recent research has shown that digital online geo- location traces are new and valuable sources to predict social interactions between users, e.g. , check-ins via FourSquare or geo-location information in Flickr images. Interestingly, if we look at related work in this area, research studying the extent to which social interactions can be predicted between users by taking more than one location-based knowledge source into account does not exist. To contribute to this field of research, we have collected social interaction data of users in an online social network called My Second Life and three related location-based knowledge sources of these users (monitored locations, shared locations and favored locations), to show the extent to which social interactions between users can be predicted. Using supervised and unsupervised machine learning techniques, we find that on the one hand the same location-based features (e.g. the common regions and common observations) perform well across the three different sources. On the other hand, we find that the shared location information is better suited to predict social interactions between users than monitored or favored location information of the user.
Social Media Mining - Chapter 10 (Behavior Analytics)SocialMediaMining
Â
R. Zafarani, M. A. Abbasi, and H. Liu, Social Media Mining: An Introduction, Cambridge University Press, 2014.
Free book and slides at http://socialmediamining.info/
NOVEL MACHINE LEARNING ALGORITHMS FOR CENTRALITY AND CLIQUES DETECTION IN YOU...ijaia
Â
The goal of this research project is to analyze the dynamics of social networks using machine learning techniques to locate maximal cliques and to find clusters for the purpose of identifying a target demographic. Unsupervised machine learning techniques are designed and implemented in this project to analyze a dataset from YouTube to discover communities in the social network and find central nodes. Different clustering algorithms are implemented and applied to the YouTube dataset. The well-known Bron-Kerbosch algorithm is used effectively in this research to find maximal cliques. The results obtained from this research could be used for advertising purposes and for building smart recommendation systems. All algorithms were implemented using Python programming language. The experimental results show that we were able to successfully find central nodes through clique-centrality and degree centrality. By utilizing clique detection algorithms, the research shown how machine learning algorithms can detect close knit groups within a larger network.
Social Media Mining - Chapter 5 (Data Mining Essentials)SocialMediaMining
Â
R. Zafarani, M. A. Abbasi, and H. Liu, Social Media Mining: An Introduction, Cambridge University Press, 2014.
Free book and slides at http://socialmediamining.info/
Research of usability of Mashup Tools done for Kent County Council as part of the Pic and Mix Pilot (2009), opening up Kent related datasets for all to use and exploit.
Identification of inference attacks on private Information from Social Networkseditorjournal
Â
Online social networks, like
Facebook, twitter are increasingly utilized by
many people. These networks permit users to
publish details about them and to connect to
their friends. Some of the details revealed
inside these networks are meant to be
keeping private. Yet it is possible to use
learning algorithms and methods on released
data have to predict private information,
which cause inference attacks. This paper
discovers how to launch inference attacks
using released social networking details to
predict private informationâs. It then
separate three possible sanitization
algorithms that could be used in various
situations. Then, it investigates the
effectiveness of these techniques and tries to
use methods of collective inference
techniques to determine sensitive attributes
of the user data set. It shows that it can
decline the effectiveness of both the local and
relational classification algorithms by using
the sanitization methods we described.
Social Targeting: Understanding Social Media Data Mining & AnalysisInfini Graph
Â
Chase McMichael â CEO, InfiniGraph
Social Targeting: Understanding Social Media Data Mining & Analysis
With the advent of the social web, companies that arenât actively mining, analysing and using social media data are missing a huge commercial advantage. In this session Chase McMichael will explain how social targeting works, including technologies, techniques and opportunities. He will also highlight the privacy challenges facing the industry.
Fuzzy AndANN Based Mining Approach Testing For Social Network AnalysisIJERA Editor
Â
Fast and Appropriate Social Network Analysis (SNA) tools ,techniques, are required to collect and classify
opinion scores on social networksites , as a grouping on wrong opinion may create problems for a society or
country . Social Network Analysis (SNA) is popular means for researcher as the number of users and groups
increasing day by day on that social sites , and a large group may influence other.In this paper, we
recommendhybrid model of opinion recommendation systems, for single user and for collective community
respectively, formed on social liking and influence network theory. By collecting thedata of user social networks
and preferenceslike, we designed aimproved hybrid prototype to imitate the social influence by like and sharing
the information among groups.The significance of this paper to analyze the suitability of ANN and Fuzzy sets
method in a hybrid manner for social web sites classifications, First, we intend to use Artificial Neural
Network(ANN)techniques in social media data classification by using some contemporary methods different
than the conventional methods of statistics and data analysis, in next we want to propagate the fuzzy approach
as a way to overcome the uncertainity that is always present in social media analysis . We give a brief overview
of the main ideas and recent results of social networks analysis , and we point to relationships between the two
social network analysis and classification approaches .This researchsuggests a hybrid classification model build
on fuzzy and artificial neural network (HFANN). Information Gain and three popular social sites are used to
collect information depicting features that are then used to train and test the proposed methods . This neoteric
approach combines the advantages of ANN and Fuzzy sets in classification accuracy with utilizing social data
and knowledge base available in the hate lexicons.
Sampling of User Behavior Using Online Social NetworkEditor IJCATR
Â
The popularity of online networks provides an opportunity to study the characteristics of online social network graphs is important, both to improve current systems and to design new application of online social networks. Although personalized search has been proposed for many years and many personalization strategies have been investigated, it is still unclear whether personalization is consistently effective on different queries for different users, and under different search contexts. In this paper, we study performance of information collection in a dynamic social network. By analyzing the results, we reveal that personalized search has significant improvement over common web search.
The mixing time of thee sampling process strongly depends on the characteristics of the graph.
R. Zafarani, M. A. Abbasi, and H. Liu, Social Media Mining: An Introduction, Cambridge University Press, 2014.
Free book and slides at http://socialmediamining.info/
Avoiding Anonymous Users in Multiple Social Media Networks (SMN)paperpublications3
Â
Abstract: The main aim of this project is secure the user login and data sharing among the social networks like Gmail, Facebook and also find anonymous user using this networks. If the original user not available in the networks, but their friends or anonymous user knows their login details means possible to misuse their chats. In this project we have to overcome the anonymous user using the network without original user knowledge. Unauthorized user using the login to chat, share images or videos etc This is the problem to be overcome in this project .That means user first register their details with one secured question and answer. Because the anonymous user can delete their chat or data In this by using the secured questions we have to recover the unauthorized user chat history or sharing details with their IP address or MAC address. So in this project they have found out a way to prevent the anonymous users misuse the original user login details.
Introduction to the Responsible Use of Social Media Monitoring and SOCMINT ToolsMike Kujawski
Â
These are my slides from a custom tool-based demonstration workshop I was asked to do where I went over various free tools that can be used to obtain valuable public data.
Social Media Mining - Chapter 6 (Community Analysis)SocialMediaMining
Â
R. Zafarani, M. A. Abbasi, and H. Liu, Social Media Mining: An Introduction, Cambridge University Press, 2014.
Free book and slides at http://socialmediamining.info/
Predicting Social Interactions from Different Sources of Location-based Knowl...Michael Steurer
Â
Recent research has shown that digital online geo- location traces are new and valuable sources to predict social interactions between users, e.g. , check-ins via FourSquare or geo-location information in Flickr images. Interestingly, if we look at related work in this area, research studying the extent to which social interactions can be predicted between users by taking more than one location-based knowledge source into account does not exist. To contribute to this field of research, we have collected social interaction data of users in an online social network called My Second Life and three related location-based knowledge sources of these users (monitored locations, shared locations and favored locations), to show the extent to which social interactions between users can be predicted. Using supervised and unsupervised machine learning techniques, we find that on the one hand the same location-based features (e.g. the common regions and common observations) perform well across the three different sources. On the other hand, we find that the shared location information is better suited to predict social interactions between users than monitored or favored location information of the user.
Social Media Mining - Chapter 10 (Behavior Analytics)SocialMediaMining
Â
R. Zafarani, M. A. Abbasi, and H. Liu, Social Media Mining: An Introduction, Cambridge University Press, 2014.
Free book and slides at http://socialmediamining.info/
NOVEL MACHINE LEARNING ALGORITHMS FOR CENTRALITY AND CLIQUES DETECTION IN YOU...ijaia
Â
The goal of this research project is to analyze the dynamics of social networks using machine learning techniques to locate maximal cliques and to find clusters for the purpose of identifying a target demographic. Unsupervised machine learning techniques are designed and implemented in this project to analyze a dataset from YouTube to discover communities in the social network and find central nodes. Different clustering algorithms are implemented and applied to the YouTube dataset. The well-known Bron-Kerbosch algorithm is used effectively in this research to find maximal cliques. The results obtained from this research could be used for advertising purposes and for building smart recommendation systems. All algorithms were implemented using Python programming language. The experimental results show that we were able to successfully find central nodes through clique-centrality and degree centrality. By utilizing clique detection algorithms, the research shown how machine learning algorithms can detect close knit groups within a larger network.
Social Media Mining - Chapter 5 (Data Mining Essentials)SocialMediaMining
Â
R. Zafarani, M. A. Abbasi, and H. Liu, Social Media Mining: An Introduction, Cambridge University Press, 2014.
Free book and slides at http://socialmediamining.info/
Research of usability of Mashup Tools done for Kent County Council as part of the Pic and Mix Pilot (2009), opening up Kent related datasets for all to use and exploit.
Identification of inference attacks on private Information from Social Networkseditorjournal
Â
Online social networks, like
Facebook, twitter are increasingly utilized by
many people. These networks permit users to
publish details about them and to connect to
their friends. Some of the details revealed
inside these networks are meant to be
keeping private. Yet it is possible to use
learning algorithms and methods on released
data have to predict private information,
which cause inference attacks. This paper
discovers how to launch inference attacks
using released social networking details to
predict private informationâs. It then
separate three possible sanitization
algorithms that could be used in various
situations. Then, it investigates the
effectiveness of these techniques and tries to
use methods of collective inference
techniques to determine sensitive attributes
of the user data set. It shows that it can
decline the effectiveness of both the local and
relational classification algorithms by using
the sanitization methods we described.
In this contribution, we develop an accurate and effective event detection method to detect events from a
Twitter stream, which uses visual and textual information to improve the performance of the mining
process. The method monitors a Twitter stream to pick up tweets having texts and images and stores them
into a database. This is followed by applying a mining algorithm to detect an event. The procedure starts
with detecting events based on text only by using the feature of the bag-of-words which is calculated using
the term frequency-inverse document frequency (TF-IDF) method. Then it detects the event based on image
only by using visual features including histogram of oriented gradients (HOG) descriptors, grey-level cooccurrence
matrix (GLCM), and color histogram. K nearest neighbours (Knn) classification is used in the
detection. The final decision of the event detection is made based on the reliabilities of text only detection
and image only detection. The experiment result showed that the proposed method achieved high accuracy
of 0.94, comparing with 0.89 with texts only, and 0.86 with images only.
Event detection in twitter using text and image fusioncsandit
Â
In this paper, we describe an accurate and effective event detection method to detect events from
Twitter stream. It detects events using visual information as well as textual information to improve
the performance of the mining. It monitors Twitter stream to pick up tweets having texts and photos
and stores them into database. Then it applies mining algorithm to detect the event. Firstly, it detects
event based on text only by using the feature of the bag-of-words which is calculated using the term
frequency-inverse document frequency (TF-IDF) method. Secondly, it detects the event based on
image only by using visual features including histogram of oriented gradients (HOG) descriptors,
grey-level co-occurrence matrix (GLCM), and color histogram. K nearest neighbours (Knn)
classification is used in the detection. Finally, the final decision of the event detection is made based
on the reliabilities of text only detection and image only detection. The experiment result showed that
the proposed method achieved high accuracy of 0.93, comparing with 0.89 with texts only, and 0.86
with images only.
IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com
IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com
Social network analysis is a method of big data analysis which reveals the nature
of connections between objects, including implicit connections. This is a tool of interest
since it can be applied to large data sets, manual processing of which is very laborintensive,
while automated processing through self-learning linguistic engines requires
a lot of resources. In this regard a study was carried out: it was aimed at development
and testing of social network analysis tools and creating a research algorithm which is
applicable to solve a wide range of analytical and search tasks. The current image of
Russia and its activities in the Arctic was chosen as a case.
The research algorithm helps to discover implicit patterns and trends, relate
information flows and events with relevant newsworthy events and news stories to form
a âclearâ view of the study object and key actors which this object is associated with.
The work contributes to filling the gap in scientific literature, caused by insufficient
development of applied issues of using social network analysis to solve managerial
tasks, while theoretical papers, which describe the theory and methodology of such an
analysis, are abundant.
For my final year project I used data analysis techniques to investigate user behavior pattern recognition in respect of similar interests and culture versus offline geographical location. This was an out-of-the-box topic, which I selected due to my love on Data Analysis, in respect of the Social Network Analysis in the Internet era.
Identity Resolution across Different Social Networks using Similarity Analysisrahulmonikasharma
Â
Today the Social Networking Sites have become very popular and are used by most of the people. This is because the Social Networking sites are playing different roles in different fields and facilitating the needs of its users from time to time. The most common purpose why people join in to these websites is to get connected with people and sharing information. An individual may be signed in on more than one Social Networking Site so identifying the same individual on different Social Networking sites is a task. To accomplish this task the proposed system uses the Similarity Analysis method on the available information details.
2010 Catalyst Conference - Trends in Social Network AnalysisMarc Smith
Â
Review of trends related to social network analysis in the enterprise. Presented at the 2010 Catalyst Conference in San Diego, CA july 29, 2010. Presented with Mike Gotta, Gartner Group.
Big Data Analytics- USE CASES SOLVED USING NETWORK ANALYSIS TECHNIQUES IN GEPHIRuchika Sharma
Â
This report is done as a part in completion of our Big Data Analysis Course at Jindal Global Business School.
In this report, we have mainly focused on literature review of 10 use-cases in the visualization task. We have worked on use cases pertaining to varied use of social media site Twitter in the political, cultural and business context; use by drug marketers and musicians among others.
SOCIAL MEDIA ANALYTICS FOR SENTIMENT ANALYSIS AND EVENT DETECTION IN SMART CI...cscpconf
Â
Smart cities utilize Internet of Things (IoT) devices and sensors to enhance the quality of the city
services including energy, transportation, health, and much more. They generate massive
volumes of structured and unstructured data on a daily basis. Also, social networks, such as
Twitter, Facebook, and Google+, are becoming a new source of real-time information in smart
cities. Social network users are acting as social sensors. These datasets so large and complex
are difficult to manage with conventional data management tools and methods. To become
valuable, this massive amount of data, known as 'big data,' needs to be processed and
comprehended to hold the promise of supporting a broad range of urban and smart cities
functions, including among others transportation, water, and energy consumption, pollution
surveillance, and smart city governance. In this work, we investigate how social media analytics
help to analyze smart city data collected from various social media sources, such as Twitter and
Facebook, to detect various events taking place in a smart city and identify the importance of
events and concerns of citizens regarding some events. A case scenario analyses the opinions of
users concerning the traffic in three largest cities in the UAE
Similar to Evolving social data mining and affective analysis (20)
GraphRAG is All You need? LLM & Knowledge GraphGuy Korland
Â
Guy Korland, CEO and Co-founder of FalkorDB, will review two articles on the integration of language models with knowledge graphs.
1. Unifying Large Language Models and Knowledge Graphs: A Roadmap.
https://arxiv.org/abs/2306.08302
2. Microsoft Research's GraphRAG paper and a review paper on various uses of knowledge graphs:
https://www.microsoft.com/en-us/research/blog/graphrag-unlocking-llm-discovery-on-narrative-private-data/
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered QualityInflectra
Â
In this insightful webinar, Inflectra explores how artificial intelligence (AI) is transforming software development and testing. Discover how AI-powered tools are revolutionizing every stage of the software development lifecycle (SDLC), from design and prototyping to testing, deployment, and monitoring.
Learn about:
âĸ The Future of Testing: How AI is shifting testing towards verification, analysis, and higher-level skills, while reducing repetitive tasks.
âĸ Test Automation: How AI-powered test case generation, optimization, and self-healing tests are making testing more efficient and effective.
âĸ Visual Testing: Explore the emerging capabilities of AI in visual testing and how it's set to revolutionize UI verification.
âĸ Inflectra's AI Solutions: See demonstrations of Inflectra's cutting-edge AI tools like the ChatGPT plugin and Azure Open AI platform, designed to streamline your testing process.
Whether you're a developer, tester, or QA professional, this webinar will give you valuable insights into how AI is shaping the future of software delivery.
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Albert Hoitingh
Â
In this session I delve into the encryption technology used in Microsoft 365 and Microsoft Purview. Including the concepts of Customer Key and Double Key Encryption.
Accelerate your Kubernetes clusters with Varnish CachingThijs Feryn
Â
A presentation about the usage and availability of Varnish on Kubernetes. This talk explores the capabilities of Varnish caching and shows how to use the Varnish Helm chart to deploy it to Kubernetes.
This presentation was delivered at K8SUG Singapore. See https://feryn.eu/presentations/accelerate-your-kubernetes-clusters-with-varnish-caching-k8sug-singapore-28-2024 for more details.
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...DanBrown980551
Â
Do you want to learn how to model and simulate an electrical network from scratch in under an hour?
Then welcome to this PowSyBl workshop, hosted by Rte, the French Transmission System Operator (TSO)!
During the webinar, you will discover the PowSyBl ecosystem as well as handle and study an electrical network through an interactive Python notebook.
PowSyBl is an open source project hosted by LF Energy, which offers a comprehensive set of features for electrical grid modelling and simulation. Among other advanced features, PowSyBl provides:
- A fully editable and extendable library for grid component modelling;
- Visualization tools to display your network;
- Grid simulation tools, such as power flows, security analyses (with or without remedial actions) and sensitivity analyses;
The framework is mostly written in Java, with a Python binding so that Python developers can access PowSyBl functionalities as well.
What you will learn during the webinar:
- For beginners: discover PowSyBl's functionalities through a quick general presentation and the notebook, without needing any expert coding skills;
- For advanced developers: master the skills to efficiently apply PowSyBl functionalities to your real-world scenarios.
Dev Dives: Train smarter, not harder â active learning and UiPath LLMs for do...UiPathCommunity
Â
đĨ Speed, accuracy, and scaling â discover the superpowers of GenAI in action with UiPath Document Understanding and Communications Miningâĸ:
See how to accelerate model training and optimize model performance with active learning
Learn about the latest enhancements to out-of-the-box document processing â with little to no training required
Get an exclusive demo of the new family of UiPath LLMs â GenAI models specialized for processing different types of documents and messages
This is a hands-on session specifically designed for automation developers and AI enthusiasts seeking to enhance their knowledge in leveraging the latest intelligent document processing capabilities offered by UiPath.
Speakers:
đ¨âđĢ Andras Palfi, Senior Product Manager, UiPath
đŠâđĢ Lenka Dulovicova, Product Program Manager, UiPath
Neuro-symbolic is not enough, we need neuro-*semantic*Frank van Harmelen
Â
Neuro-symbolic (NeSy) AI is on the rise. However, simply machine learning on just any symbolic structure is not sufficient to really harvest the gains of NeSy. These will only be gained when the symbolic structures have an actual semantics. I give an operational definition of semantics as âpredictable inferenceâ.
All of this illustrated with link prediction over knowledge graphs, but the argument is general.
Connector Corner: Automate dynamic content and events by pushing a buttonDianaGray10
Â
Here is something new! In our next Connector Corner webinar, we will demonstrate how you can use a single workflow to:
Create a campaign using Mailchimp with merge tags/fields
Send an interactive Slack channel message (using buttons)
Have the message received by managers and peers along with a test email for review
But thereâs more:
In a second workflow supporting the same use case, youâll see:
Your campaign sent to target colleagues for approval
If the âApproveâ button is clicked, a Jira/Zendesk ticket is created for the marketing design team
Butâif the âRejectâ button is pushed, colleagues will be alerted via Slack message
Join us to learn more about this new, human-in-the-loop capability, brought to you by Integration Service connectors.
And...
Speakers:
Akshay Agnihotri, Product Manager
Charlie Greenberg, Host
Essentials of Automations: Optimizing FME Workflows with ParametersSafe Software
Â
Are you looking to streamline your workflows and boost your projectsâ efficiency? Do you find yourself searching for ways to add flexibility and control over your FME workflows? If so, youâre in the right place.
Join us for an insightful dive into the world of FME parameters, a critical element in optimizing workflow efficiency. This webinar marks the beginning of our three-part âEssentials of Automationâ series. This first webinar is designed to equip you with the knowledge and skills to utilize parameters effectively: enhancing the flexibility, maintainability, and user control of your FME projects.
Hereâs what youâll gain:
- Essentials of FME Parameters: Understand the pivotal role of parameters, including Reader/Writer, Transformer, User, and FME Flow categories. Discover how they are the key to unlocking automation and optimization within your workflows.
- Practical Applications in FME Form: Delve into key user parameter types including choice, connections, and file URLs. Allow users to control how a workflow runs, making your workflows more reusable. Learn to import values and deliver the best user experience for your workflows while enhancing accuracy.
- Optimization Strategies in FME Flow: Explore the creation and strategic deployment of parameters in FME Flow, including the use of deployment and geometry parameters, to maximize workflow efficiency.
- Pro Tips for Success: Gain insights on parameterizing connections and leveraging new features like Conditional Visibility for clarity and simplicity.
Weâll wrap up with a glimpse into future webinars, followed by a Q&A session to address your specific questions surrounding this topic.
Donât miss this opportunity to elevate your FME expertise and drive your projects to new heights of efficiency.
DevOps and Testing slides at DASA ConnectKari Kakkonen
Â
My and Rik Marselis slides at 30.5.2024 DASA Connect conference. We discuss about what is testing, then what is agile testing and finally what is Testing in DevOps. Finally we had lovely workshop with the participants trying to find out different ways to think about quality and testing in different parts of the DevOps infinity loop.
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
Â
Evolving social data mining and affective analysis
1. EVOLVING SOCIAL DATA MINING AND
AFFECTIVE ANALYSIS
METHODOLOGIES, FRAMEWORKAND
APPLICATIONS
Prof. Athena Vakali
IDEAS 2012, Prague, August 8th
2012
2. Presentation Outline
īą Web 2.0 facts and social data
īą Social associations and all kinds of graphs
īą Evolving social data mining
īą Emotion-aware social data analysis
īą Frameworks and Applications
3. īą Web 2.0 facts and social data
īą evolution & characteristics
īą is there hidden information there?
īą motivation for social (evolving) data mining
īą Social associations and all kinds of graphs
īą Emotion-aware social data analysis
īą Frameworks and Applications
4. īą Web 2.0 has become a source of vast amounts of social data
evolving at fast rates
īą Users participate massively in Web 2.0 applications such as:
īą social networking sites (e.g. )
īą blogs, microblogs (e.g. )
īą social bookmarking/tagging systems (e.g.
)
īą Social data refer to different types of entities
īą users
īą content handled separately or together
īą metadata
associated via various types of interactions /relationships
Web 2.0 facts & social data
5. SOCIAL DATA SOURCES: is there hidden
information here?
īą Web 2.0 users act explicitly by declaring their associations
īą Relationships may be multipartite (e.g. user u1 making a
comment on the post p of user u2) but are usually
simplified in bipartite associations between some involved
entities
īą However, explicit associations generate implicit threads of
relationships which:
īą are triggered by usersâ common activities
īą may hold among the non-user entities (resources,
user-user
âis frie nd withâ
user-user
âis frie nd withâ
user-resource
âlike sâ
user-resource
âlike sâ
user-metadata
âcre ate s tag â
user-metadata
âcre ate s tag â
user-group
âbe lo ng s to â
user-group
âbe lo ng s to â
6. Web 2.0 relations initiated by usersâ online
behavior
IRinference
mechanism
7. Implicit relations uncovered for various types of
entities
īą Numerous relations might unfold âĻ however only some of them are
selected based on analysisâ focus or applicationâs context
īą user-user IRs such as: âlike sam e re so urce â, âuse sam e tag â can be
leveraged for studying behavioral patterns (e.g. in applications like
Flickr)
īą tag-tag IRs such as: âare assig ne d o n sam e re so urce â can be
leveraged for tag clustering
âĻ g e ne ratio n m e chanism s
8. Motivation forSocial Data
Miningīą The availability of massive sizes of data gave new impetus to
data mining.
īą e.g. more than active 500 million active Facebookusers,
sharing on average more than 30 billion pieces of content
each month [Facebook Statistics 2011]
īą Mining social web data can act as a barometer of the usersâ
opinion. Non-obvious results may emerge.
īą Collaboration and contribution of many individuals
formation of collectiveintelligence
īą Wisdomof thecrowd: more accurate, unbiased source of
information.
īą Social data mining results can be useful for applications such
as recommender systems, automatic event detectors, etc
9. Static vs evolving data mining
īą Social data interactions are constantly updating in fast
rates
īą however, a data mining approach could be static or evolving
static mining approaches: aggregate
all social interactions over a specific
period and deal with them as a unique
dataset
evolving mining approaches:
track and exploit more fine-grained & âricherâ information
dynamic data mining:
īŧemphasis placed on data
evolution and not on
aggregation
īŧ data modeling with a given
time granularity which affects
the amount of details
contained in the dataset
streaming data mining:
īŧnew user activity data received in a
streaming fashion
īŧtime-aware data approximating
model incrementally created and
maintained, subject to time and
space constraints
īŧmodel readapted on arrival of either
single update or batch of updates
10. Motivation forEvolving Social Data
Mining
īą Identifying over time the events that affect social
interactions
īŧ tracking posts in a micro-blogging website to identify
floods, fires, riots, or other events and inform the public
īą Highlighting trends in usersâ opinions, preferences, etc
īŧ companies can track customersâ opinions and complaints
in a timely fashion to make strategic decisions
īą Tracking the evolution of groups (co m m unitie s) of
users or resources, finding changes in time and
correlations
īŧ develop better personalized recommender systems to
improve user experience
11. īą Social data in the Web 2.0
īą Social associations and all kinds of graphs
īą structures for static social data
īą evolving data representation structures
īą Evolving social data mining
īą Emotion-aware social data analysis
īą Frameworks and Applications
12. The ne two rk m o de las an o bvio us cho ice âĻ
īąSocial data are interconnected through associations
forming a networkor graph G(V, E), where Vis the set of
nodes and E is the set of edges.
īą nodes represent entities/objects and edges
represent relations
īą different types of nodes and edges
īą weighted/unweighted
īą directed/undirected
Social associations and all kinds of
graphs
14. Structures forstatic social data
īą Hypergraph: generalization of a graph where an
edge (hype re dg e ) connects more than two nodes
[Brinkmeier07]
īą Folksonomy: lightweight knowledge representation
emerging from the use of a shared vocabulary to
characterize resources â tripartite hype rg raph
[Hotho06, Mika05]
īą Projection on simple graphs to lower complexity [Au
Yeung09]
īą further simplifications in bipartite & unipartite g raphs
īą e.g. tag-tag network where two tags are connected if
assigned to the same resource
īą Simple graphsâ structure can be encoded in an
tripartite graph
16. Evolving data representation
structures
īą Need for modeling the different data states in successive time-
steps, often determined by the dataâs sampling rate
t1 t2 t3 t4 âĻ tk-1 tk
G1 G2 G3 G4 âĻ Gk-1 Gk
seg1 seg2 âĻ segm
the graph stream as
a sequence of
snapshot graphs: G
= {Gt }, tââ
G
graph
snapshots
timeline
segments
graph stream
17. t1 t2 t3 t4 âĻ tk-1 tk
G1 G2 G3 G4 âĻ Gk-1 Gk
seg1 seg2 âĻ segm
The snapshot layer
G
adjacency/similarity
matrices
folksonomies
graph
snapshots
timeline
segments
graph streamthe graph stream as
a sequence of
snapshot graphs: G
= {Gt }, tââ
18. t1 t2 t3 t4 âĻ tk-1 tk
G1 G2 G3 G4 âĻ Gk-1 Gk
seg1 seg2 âĻ segm
The segment layer
G
graph
snapshots
timeline
segments
graph streamthe graph stream as
a sequence of
snapshot graphs: G
= {Gt }, tââ
Pre-processing technique
Identify graph segments consisting of
similar snapshots and compute a
smooth graph approximation for each
segment [Yang09]
adjacency/similarity matrices
īŧmore coarse-grained
īŧtime-aggregated matrices are
used for emphasizing on most
recent edges [Tong08]
tensors: generalization of matrices (> 2
dimensions) [Sun06]
19. t1 t2 t3 t4 âĻ tk-1 tk
G1 G2 G3 G4 âĻ Gk-1 Gk
seg1 seg2 âĻ segm
The streamlayer
G
graph
snapshots
timeline
segments
graph streamthe graph stream as
a sequence of
snapshot graphs: G
= {Gt }, tââ tensors
âmulti-graphsâ with edges
encoding relations as well as
temporal information [Zhao07]
time aggregate adjacency
matrices [Tong08]
20. References
[Giatsoglou12] M. Giatsoglou, A. Vakali. Capturing Social Data Evolution via Graph Clustering. In
IEEE Internet Computing, IEEE computer Society Digital Library. IEEE Computer Society. DOI:
10.1109/MIC.2012.24, Feb. 2012.
Au Yeung09] Au Yeung, C.M., Gibbins, N., and Shadbolt, N. 2009. Contextualising Tags in
Collaborative Tagging Systems. In Proceedings of 20th ACM Conference on Hypertext and
Hypermedia, pp. 251-260.
[Brinkmeier07] Brinkmeier, M., Werner, J., and Recknagel, S. 2007. Communities in graphs and
hypergraphs. CIKM'07. ACM, 869-872.
[Hotho06] Hotho, A., Robert, J., Christoph, S., and Gerd, S. 2006. Emergent Semantics in
BibSonomy. GI Jahrestagung Vol. P-94, 305â312. Gesellschaft fr Informatik.
[Mika05] Mika, P. 2005. Ontologies Are Us: A Unified Model of Social Networks and Semantics. In
Proceedings of the 4th international SemanticWeb Conference. ISWCâ05. Springer Berlin
/Heidelberg, pp. 522-536.
[Sun07] Sun, J., Faloutsos, C., Papadimitriou, S., and Yu, P. S. 2007. GraphScope: parameter-free
mining of large time-evolving graphs. In Proceedings of the 13th ACM SIGKDD international
Conference on Knowledge Discovery and Data Mining . KDD '07. ACM, 687-696.
[Tong08] Tong, H., Papadimitriou, S., Yu, P. S., and Faloutsos, C. 2008. Proximity tracking on time-
evolving bipartite graphs. In Proceedings of the 9th SIAM international Conference on Data Mining .
21. Presentation Outline
īą Social data in the Web 2.0
īą Social associations and all kinds of graphs
īą Evolving social data mining
īą the clustering approach
īą Applications
īą Emotion-aware social data analysis
īą Frameworks and applications
22. Web 2.0 social data mining : The generic
workflow
next, focus on clustering and community detection
23. Evolving social data clustering/community detection
approaches
Preliminary efforts followed the Community
Mapping -CM- approach:
1.application of traditional community detection algorithms
on individual static graph snapshots
2.identification of correlations and mapping between
successive snapshotsâ communities using special
similarity measures & temporal smoothing techniques
Limitation: tend to find community structures with
high temporal variation (due to real world
ambiguous & noisy data)
24. Evolving social data clustering/community detection
approaches
Evolutionary community identification approaches: utilize
community structureâs history to maximize temporal
smoothness and lead to smoother community evolution
īą
traditional clustering revisited foran evolutionary setting (TCR
approach): modification of existing static clustering algorithms to have
memory of previous states of the data;
īą
spectral clustering (SC approach): uses the spectrum of the graphâs
similarity matrix to perform dimensionality reduction for clustering in fewer
dimensions;
īą
non-negative matrix/tensorfactorization (NFC approach): apply NFC for
discovering communities jointly maximizing the fit to the observed data &
the temporal evolution;
īągraph streamsegment identification & community structure detection
25. CM approach
Palla
2008
undirecte
d graph
snapshot
s
I. co-
authorship
network
II. mobile
phone call
network
I. 142 months
II. 52 weeks
(26
tim e slo ts)
I. {30K authors}
{âare co -autho rs â }
II. {4M users} {âcalls â}
community detection
based on cliq ue
pe rco latio n;
successive
communities mapping
by relative nodesâ
overlap;
Lin
2007
directed
graph
snapshot
s
data crawled
from 407 blogs
63 weeks {275K bloggers}
{149K âre plie s to po st
o fâ}
hypothesis: use rsâ
mutualawareness (e . g .
blo g g e rs co m m e nting
o n e ach o the râs blo g )
drive s co m m unity
fo rm atio n;
mutual awarenessâ
expansion as a
random walk process
leads to community
detection
method structure
dataset
outcome/scope
source time-period entities / relations
26. TCRapproach
Chakrabar
ti
2006
(time-
aware)
similarity
matrix
photo
sharing
service
68 weeks {5K tags}
{âare applie d o n sam e
re so urce â}
joint
optimization of
2 criteria:
ī§snapshot
quality
ī§history quality;
generic
framework with
2 instantiations
proposed:
ī§agglomerative
hierarchical
algorithm
ī§K-means
method structure
dataset
outcome/scope
source time-period entities / relations
Simultaneous optimization of two potentially conflicting criteria:
(i) snapshot quality sq, and (ii) history quality hq
At each time-step the framework finds a clustering based on the new similarity matrix Mt
and the so far history
Evolutionary clustering in an online setting
27. SC approach
Tang
2008
series of
network
snapshots;
interaction
matrix for each
snapshot
I. mail
exchange
network
II. co-authorship
network
I. 12
months
II. 25 years
I. {2.4K users}
{emails}
{36.7K words}
{âse nds
e m ailâ}
{âre ce ive s
e m ailâ}
{âco ntains
te rm â}
II. {492K papers}
{347K authors}
{2.8K venues}
{9.5K title
terms} {âwrite sâ}
{âparticipate s
inâ} {âis
publishe d inâ}
community
evolution in
dynamic networks
of multiple social
entities;
iterative
approximation of
community
evolution using
eigenvector
calculation and
K-means clustering
method
structure or
model
dataset
outcome/scopesource period entities / relations
Spectral clustering uses the spectrum of the
graphâs similarity matrix to
perform dimensionality reduction for clustering
in fewer dimensions
Evolutionary spectral clustering applied
in multi-mode networks
28. NFC approach
Lin
2009
metagraph:
hypergraph with
nodes
representing
face ts and edges
multipartite
interactions;
tensors
social
bookmarkin
g service
with voting
capabilities
27 days
(9
tim e slo ts)
{users} {posts} {keywords}
{topics}
{152K âpo st-ke ywo rd-
to picâ}
{56.4K âis frie nd withâ}
{44Kâuplo ads â}
{1.2M âvo te s o nâ}
{242K âuse r-po st-
co m m e ntâ}
{94.6K âre plie s withâ}
community
extraction via time-
stamped tensor
factorization;
on-line method
handling time-
varying relations
through incremental
metagraph
factorization;
communities derived
by jointly leveraging
all types of
multipartite relations
method structure or model
dataset
outcome/scope
source period entities / relations
29. SGC approach
Sun
2007
unweighted
undirected
bipartite
graphs;
graph
segments
I. mail
exchange
network
II. mobile phone
call network
III. mobile
device
proximity
records
I. 165
weeks
II. 46 weeks
III. 46 weeks
I. {34.3K senders}
{34.3K recipients}
{15K
âse nd m ailâ / week}
II. {96 callers} {3.8K
callees}
{ 430
âcallsâ/ week}
III. {96 users } {96 users}
{689 âis lo cate d
ne arâ/week}
unparametric
method based
on Minimum
Description
Length;
each
segmentâs
source and
destination
nodes are
partitioned
separately to
decrease cost;
compressed
graph in < 4%
than original
space
method structure
dataset
outcome/scope
source period entities / relations
30. References (1/2)
[Chakrabarti06] Chakrabarti, D., Kumar, R., and Tomkins, A. 2006. Evolutionary clustering. In Proc. of
KDD '06. ACM, New York, NY, 554-560.
[Tang08] Tang, L., Liu, H., Zhang, J., and Nazeri, Z. 2008. Community evolution in dynamic multi-mode
networks. In Proc.. KDD '08. ACM, New York, NY, 677-685.
[Palla05] Palla, G., DerÊny, I., Farkas, I., and Vicsek, T. 2005. Uncovering the overlapping community
structure of complex networks in nature and society. Nature 435, 814â818.
[Palla07] Palla, G., BarabÃĄsi, A.-L., and Vicsek, T. 2007. Quantifying social group evolution. Nature
446, 664-667.
[Lin07] Lin, Y., Sundaram, H., Chi, Y., Tatemura, J., and Tseng, B. L. 2007. Blog Community Discovery
and Evolution Based on Mutual Awareness Expansion. In Proceedings of the IEEE/WIC/ACM
international Conference on Web intelligence. Web Intelligence. IEEE Computer Society, Washington,
DC, 48-56.
[Lin09] Lin, Y., Sun, J., Castro, P., Konuru, R., Sundaram, H., and Kelliher, A. 2009. MetaFac:
community discovery via relational hypergraph factorization. KDD'09. ACM, 527-536.
[Sun07] Sun, J., Faloutsos, C., Papadimitriou, S., and Yu, P. S. 2007. GraphScope: parameter-free
mining of large time-evolving graphs. In Proc. of KDD '07. ACM,, 687-696.
[Quack08] Quack, T., Leibe, B., and Van Gool, L. 2008. World-scale mining of objects and events from
community photo collections. In Proc. of CIVR '08. ACM, 47-56.
31. References (2/2)
[Zhao07] Zhao, Q., Mitra, P., and Chen, B. 2007. Temporal and information flow based event detection
from social text streams. In Proc. of the 22nd National Conference on Artificial intelligence - Volume 2.
Aaai Conference On Artificial Intelligence. AAAI Press, 1501-1506.
[Duan09] Duan, D., Li, Y., Jin, Y., and Lu, Z. 2009. Community mining on weighted directed graphs.
In Proc. of CNIKM '09. ACM, New York, NY, 11-18.
[Chi06] Chi, Y., Tseng, B. L., and Tatemura, J. 2006. Eigen-trend: trend analysis in the blogosphere
based on singular value decompositions. In Proc of CIKM '06. ACM, 68-77.
[Papadopoulos11] Papadopoulos, S., Zigkolis, C., Kompatsiaris, Y., and Vakali A. 2011. Cluster-Based
Landmark and Event Detection for Tagged Photo Collections. IEEE Multimedia, pp. 52-63.
[Java07] Java, A., Song, X., Finin, T., and Tseng, B. 2007. Why we twitter: understanding microblogging
usage and communities. WebKDD/SNA-KDD'07. ACM, 56-65.
[Jansen,09] Jansen, B. J., Zhang, M., Sobel, K., and Chowdury, A. 2009. Twitter power: Tweets as
electronic word of mouth. J. Am. Soc. Inf. Sci. Technol. 60, 11, 2169-2188.
[Sankaranarayanan 09] Sankaranarayanan, J., Samet, H., Teitler, B.E., Lieberman, M.D., and Sperling,
J. 2009. TwitterStand: news in tweets. GIS'09. ACM, 42-51.
32. Why focusing on time as a
criterion ?
ī¨ typical analysis involves âstaticâ
views (users-tags)
ī¨ events, trends affect user interests
ī¨ users Tagging Behavior changes
over time
ī¨ Time is a fundamental dimension
in analysis of users and tags in a
social tagging system
e.g. : prediction of first weekend box-
office revenues using tweets
33. Many times, a userâs targeted interest is
hidden in the general tagging activityâĻ.
Accessor
ies,
bags,
fashion,
Cars, football,
holidays, horses,
sea, turkey, fashion
New York, hat,
trousers,
fashion, Gucci
animals, elephants,
nature sea, turkey,
bags
hats,
Gucci
fashion,
jeans, NY
User 1 User 2 User 3
fashionweek,
fashion, silk,
wool
35. Related Approaches
ī¨ Sun and colleagues [Sun08] use the Ī2
statistical model, to
determine whether the appearance of tag t in a tim e fram e i
is sig nificant and, thus, to disco ve r tag s that co nstitute
âtopics of interest" at particular time frames.
ī¨ Wetzker and colleagues [Wetzker08] claim that a tag
signifies a trend, if it attracts significantly more new users in
a currently monitored time frame than in past time frames.
ī¨ A trend detection measure is introduced in [Hotho06], which
captures topic-specific trends at each time frame and is
based on the weight-spreading ranking of the PageRank
algorithm
36. Use Cases
ī¨ Capturing trends, interests, periodic activities
of users in specific time periods
ī¨ Community-based tag recommendation
ī¨ Personalization (time-aware user profiles)
ī¨ Fighting spam on social web sites (by
discriminating regular and occasional users)
37. īą Social data in the Web 2.0
īą Social associations and all kinds of graphs
īą Evolving social data mining
īą In & out zooming on time aware user/tag clusters
īą Emotion-aware social data analysis
īą User generated data
īą The idea for emotion aware clustering
īą Application on micro-blogging sources
īą Frameworks and applications
38. The idea for emotion-aware
analysis
Clustering methods embed various criteria such as:
semantics, tags, time, geo-information ... etc
but
since social sources are driven and managed by
humans
īemotion/sentiment
MUST also be considered âĻ
Clustering methods embed various criteria such as:
semantics, tags, time, geo-information ... etc
but
since social sources are driven and managed by
humans
īemotion/sentiment
MUST also be considered âĻ
39. The nature of evolving social
data
ī¨ Two main types of textual information.
ī¤ Facts and Opinions
īŽ Note: factual statements can imply opinions too.
ī¨ Most current text information processing
methods (e.g., web search, text mining) work
with factual information.
ī¨ Sentiment analysis or opinion mining
ī¤ computational study of opinions, sentiments and
emotions expressed in text.
ī¨ Why sentiment analysis now? Mainly because
of the Web; huge volumes of opinionated text.
40. User-generated data (I)
Opinions are important because whenever we need to make
a decision, weâre influenced by othersâ opinions.
ī¤ According to [Horrigan09] of more than 2000 American adults
:
īŽ 81% of Internet users (or 60% of Americans) have done online
research on a product at least once;
īŽ among readers of online reviews of restaurants, hotels, and
various services (e.g., travel agencies or doctors), between 73%
and 87% report that reviews had a significant influence on their
purchase;
īŽ 32% have provided a rating on a product, service, or person via
an online ratings system, and 30% have posted an online
comment or review regarding a product or service
41. user-generated data (II)
ī¨ Word-of-mouth on the Web
ī¤ User-generated media: One can express opinions on anything in
reviews, forums, discussion groups, blogs ...
ī¤ Opinions of global scale: No longer limited to:
īŽ Individuals: oneâs circle of friends
īŽ Businesses: Small scale surveys, tiny focus groups, etc.
ī¨ Why affect/sentiment analysis?
īŽ Customers: need peer opinions to make purchase decisions
īŽ Business providers:
īŽ need customersâ opinions to improve product
īŽ need to track opinions to make marketing decisions
īŽ Social researchers: want to know peopleâs reactions about
social events
īŽ Government: wants to know peopleâs reactions to a new policy
īŽ Psychology, education, etc.
42. More Applications
ī¨ Product review mining: What features of the iPhone
do customers like and which do they dislike?
ī¨ Review classification: Is a review positive or negative
toward the iPhone?
ī¨ Tracking sentiments toward topics overtime: Is anger
ratcheting up or cooling down?
ī¨ Prediction (election outcomes, market trends): Will
Clinton or Obama win?
43. Why Extracting sentiments from Web 2.0
data sources?
ī¨ Web 2.0 data features
ī¤ Easy to collect: huge amount, clean format
ī¤ Broadly distributed: demographics
ī¤ Topic diversified: free discussion about any topic/product/event
ī¤ Opinion rich: highly personalized
ī¤ Distributed over time, user generated content
ī¨ Motivation
ī¤ Sentiment is a very natural expression of a human being.
ī¤ Sentiment Analysis aims at getting sentiment-related knowledge
especially from the huge amount of information on the internet
ī¤ Can be generally used to understand opinion in a set of
documents or user generated content
44. Challenges
ī¨ Contrasts with Standard Fact-Based Textual Analysis
ī¤ typically, text categorization seeks to classify documents by
topic
ī¤ BUT nature, strength of feelings, degree of positivity, etc
imposes a tailored sentiment categorization
ī¨ Key Factors that Make Sentiment Analysis challenging
ī¤ choosing the right set of keywords might be less trivial than
one might initially think;
ī¤ Sentiment and subjectivity are quite context-sensitive, and, at
a coarser granularity, quite domain dependent
īŽ e.g., âgo read the bookâ most likely indicates positive sentiment for
book reviews, but negative sentiment for movie reviews.
ī¤ Web users postings are of a challenging nature, since there is
no code in expressions
45. so far âĻ Lexical Resources
ī¨ Se ntiWo rdne t
ī¨ Built on the top of WordNet synsets
ī¨ Attaches sentiment-related information with synsets
ī¨ SentiWordNet assigns to each synset of WordNet three
sentiment scores: positivity, negativity, objectivity
ī¨ Ge ne ralInq uire r
ī¤ Included are manually-classified terms labeled with various
types of positive or negative semantic orientation, and
words having to do with agreement or disagreement.
ī¨ O pinio nFinde râs Subje ctivity Le xico n
ī¤ OpinionFinder is a system that performs subje ctivity
analysis , automatically identifying when opinions,
sentiments, speculations, and other private state s are
present in text.
46. so far âĻ Lydia System
ī¨ Lydia [Lioyd05] news analysis system does a
daily analysis of over 1000+ online English
newspapers, Blogs, RSS feeds, and other news
sources.
ī¨ It identifies who is being talked about, by whom,
when and where?
ī¨ Applications of Lydia
ī¤ heatmap generation (pos/neg
for a topic);
ī¤ relational networks
47. so far âĻ integration of news &
blogs
ī¨ Bautin, Vijayarenu and Skiena [Bautin08]
presented an approach for the international
analysis for news and blogs âĻ still on the
positive/negative side âĻ
ī¤ Cross-language analysis across news streams
Polarity score of London in Arabic,
German, Italian and Spanish over the May
1-10, 2007 period.
48. so far âĻ sentiment-aware
searching
ī¨ Sentiment Analysis for
Semantic Enrichment of
Web Search Results
[Demartini10]
ī¨ the first few results a
representative sample of
the entire result set
Average Sentiment score in top N results for 3 search engines
49. so far âĻ beyond pos/neg : the affect
analysis
ī¨ it involves several affects at the same
time.
ī¨ affect classes may be correlated or
opposed.
ī¨ Abbasi, Chen Thoms and Fu
[Abbasi08] proposed a support vector
regression correlation ensemble
(SVRCE) method for text-based affect
classification.
ī¤ affect feature and technique
comparison.
ī¤ apply to multi-domain.
I cannot stand you!
hate angry
I cannot stand
you!
Affect class Intensity
Happiness 0.01
Sadness 0.03
Anger 0.6
Hate 0.5
50. īą Social data in the Web 2.0
īą Social associations and all kinds of graphs
īą Evolving social data mining
īą In & out zooming on time aware user/tag clusters
īą Emotion-aware social data analysis
īą Frameworks and applications
īą Most popular applications
īą A mining and analysis framework
īą Social data analysis on the cloud
īą Emotions capturing in microblogs
51. clustering of users
exploiting the time
dimension
Applications of Mining Evolving Social
DataThe results of community detection, or different mining
techniques, on evolving social data can be exploited in
applications:
social network analysis
im ag e from [Touchgraph]
event detection
diag ram fro m [Sun07]
trend detection
im a g e from [Trendsmap]
[Touchgraph] http: //www. to uchg raph. co m /TG Face bo o kBro wse r. htm l
[Trendsmap] http: //tre ndsm ap. co m /
52. Event detection
Graph segmentation
community detection
methods [Sun07, Duan09]
identify events as
significant change-
timepoints in the
stream.
Definition of event
ī§information flow between a group of social
actors on a specific topic over a certain time
period [Zhao07]
ī§occasions which take place at a specific
time and location (concerts, festivals, etc.)
[Quack08]
Event & landmarkdetection from geo-tagged
data [Papadopoulos11]
īądetection of image clusters using visual & tag-
based similarities
īąclassification of clusters as events or landmarks
īą Landmarks differ from events on the number of
users tagging photos over different time
granularities
īą exploitation of similarities with disjoint sets of
landmark- and event- related tags
event detection fromsocial data
streams [Zhao07]
īąfeatures exploration in 3
dimensions: textual content, social,
temporal
īągeneration of multiple intermediate
clustering structures using content-
based similarity & information flow
patterns
īąevents as groups of nodes closely
related in time & topic
event detection from social data
streams [Zhao07]
īąfeatures exploration in 3
dimensions: textual content, social,
temporal
īągeneration of multiple intermediate
clustering structures using content-
based similarity & information flow
patterns
īąevents as groups of nodes closely
related in time & topic
53. Trend tracking
īą Social data fluctuate in structure and frequency as they evolve and over time some
topics, images, tags, etc, become most popular amongst users.
Trends can be identified by a
data mining approach g lo bally or lo cally (within communities) and they usually
indicate what interests users the most at a given time.
trend detection in blogs [Chi06]
īą focused on:
īŧ keyword popularity in successive
timeframes
īŧ detection of different topics relating to a
keyword
īŧ contribution of individual users to a trend
īą uses the results of SingularValue
Decomposition as trend indicators
capturing both temporal data
changes & bloggersâ characteristics
īą exploits textual content and citations
between blogs
trend detection in Twitter[Java07],
[Jansen09],
[Sankaranarayanan09]
īą several attempts using statistical
analysis methods
īŧ analysis of evolving Twitter data to
identify trending keywords for
different weekdays
īŧ sentiment identification on tweets
to identify trending sentiments
about brands
īą online clustering on streaming
tweets, combined with
classification, to identify breaking
news
trend detection in Twitter[Java07],
[Jansen09],
[Sankaranarayanan09]
īą several attempts using statistical
analysis methods
īŧ analysis of evolving Twitter data to
identify trending keywords for
different weekdays
īŧ sentiment identification on tweets
to identify trending sentiments
about brands
īą online clustering on streaming
tweets, combined with
classification, to identify breaking
news
55. 2 indicative applicatio ns
ī¨ Cloud4Trends
Leveraging the cloud infrastructure for
localized real-time trend detection in social
media
ī¨ CapturEmos
Capturing emotional patterns in micro-
blogging data streams
Aristotle University, OSWINDS Group
56. Cloud4Trends is a microblogging & blogging localized content
collection and analysis frameworkfor detecting currently popular
topics of usersâ interest
Cloud4Trends is a microblogging & blogging localized content
collection and analysis frameworkfor detecting currently popular
topics of usersâ interest
ī¨ Social media reflect societal concerns exhibiting
âburstsâ of content generation on the occurrence of
events
ī¤ po pular to pics / inte re sts fluctuate with tim e
ī¨ Challenging for both computer scientists &
application developers to reach unbiased,
meaningful conclusions about trendingusersâ
opinion and interests
Cloud4Trends - Motivation
57. Challenges & Outcome
ī¨ Massive content sizes and
unpredictable content generation
rates :
ī¤ scalable analysis is needed
ī¨ Trending topics should be
discovered when they are âfreshâ :
ī¤ an on-line analysis approach is
demanded
ī¨ Trends should be meaningful
ī¤ need for contextual trends
ī¨ Content is dispersed in multiple
sources :
ī¤ trend detection needs a combined
approach
The Cloud deployment of
the Cloud4Trends
scenario with use of the
VENUS-C services
verified that Cloud-based
architectures are a viable
solution for online web
data mining applications
that are beneficial for
both researchers and
entrepreneurs.
59. Emotional Aware Clustering on Micro-
Blogging Sources (affect analysis)
K. Tsagkalidou, V. Koutsonikola, A.
Vakali and. K. Kafetsios: Emotional
Aware Clustering on Micro-Blogging
Sources, accepted for publication,
Affective Computing and Intelligent
Interaction 2011
60. our emotional dictionary
ī¨ create an extended emotional dictionary by
enriching an opinion lexicon provided by the
UMBC university with synonymous words from
WordNet
61. The used affect space
ī¨ representing the extreme ends of four emotional pairs
[Gill08]
ī¨ emotion exemplar words
62. Relevance between tweet and
emotion
ī¨ Semantic similarity
ī¤ the maximum semantic similarity between each
of the tweetâs wo rds and the emotionâs
representatives, as defined in Wordnet
ī¨ Sentiment similarity
ī¤ expresses a wordâs emotional intensity as defined
in the extended dictionary
ī¨ Overall similarity between a tweet & an
emotion ÎŖi=1âĻ|words|
Sem(wi,emotion)*Sent(wi)
|wi|: Sent(wi)â 0, for 1<=i<=|words|
63. âĸ capturing and understanding crowdâs emotions
for a particular topic or product in an implicit
manner via computational methods.
âĸ se ntim e nt analysis & m icro blo g g ing (statistical)
pro ce ssing : emphasis on affective and opinion
mining, lexicon-based processing, knowledge
extraction techniques;
âĸ de ve lo pm e nt o f applicatio ns: web application
enhanced with of crowds emotions visualization
capabilities.
CapturEmos- MotivationCapturEmos- Motivation
64. The application/service
useful for capturing branding
success & diffusion in the
market, as expressed by the
crowds emotions
Our innovation principle :
focus on the âaffectâ which is
distinguished from discrete
emotions.
discrete emotions : concern
affective reactions in relation
to oneâs goals
affect refers to : an
overarching positive or
negative valence of oneâs
feelings.
http://oswinds2.csd.auth.gr/~vkoutson/EmoGlobe/
65. Niche targeted Markets
political stakeholders, public authorities (e.g. municipalities),
consumer behavior policy makers, chambers of commerce,
tourism and infotainment sectorsâĻ
markets characteristics : emerging, unpredicted bursts, multi-
profiles
Challenges : multi-lingual support; privacy and anonymity
preservation;
real-time emerging data flows; time and space complexities;
Markets & challenges
proposed framework and applications :
âĸ address wide stakeholders and markets audiences;
âĸ certain tasks can be realized (e.g. capturing branding success &
diffusion in the market) expressed by the crowds emotions;
âĸ can support policy and decision making.
66. Future work and horizon ...
ī¨ emerging, and unpredicted bursts detections
in evolving social media;
ī¨ user multi-profiles patterns;
ī¨ support applications with multi-lingual support;
ī¨ privacy and anonymity preservation
ī¨ development of intelligent and collective
information retrieval techniques are required
and well expected.
67. References
[Horrigan09] J. A. Horrigan, âOnline shopping,â Pew Internet & American Life Project Report, 2008.
[comScore07] comScore/the Kelsey group, âOnline consumer-generated reviews have significant impact on offline
purchase behavior,â Press Release, http://www.comscore.com/press/release.asp?press=1928, November 2007.
[Lioyd05] Lloyd, L., Kechagias, D., Skiena, S.: Lydia: A system for large-scale news analysis. In: String Processing
and Information Retrieval (SPIRE 2005). Volume Lecture Notes in Computer Science, 3772. (2005) 161-166
[Bautin08] M. Bautin, L. Vijayarenu, S. Skiena : International sentiment analysis for News and Blogs, In Proceedings
of ICWSM (2008)
[Demartini10] G. Demartini and S. Siersdorfer: Dear Search Engine: Whatâs youropinion about...? Sentiment Analysis
for Semantic Enrichment of Web Search Results, In Proc. Of WWW201 0, April 2630, 2010
[Abbasi08] A. Abbasi, H. Chen, S. Thoms, T. Fu: Affect Analysis of Web Forums and Blogs Using Correlation
Ensembles, IEEE Transactions on Knowledge and Data Engineering (2008) Vol. 20, Issue 9
[OM] Opinion Mining and Sentiment Analysis: NLP Meets Social Sciences, Bing Liu Department of Computer Science
University Of Illinois at Chicago
[Sun 08] Sun A, Zeng D, Li H, Zheng X (2008) Discovering trends in collaborative tagging systems. In: Proceedings of
the IEEE ISI 2008 PAISI, PACCF, and SOCO international workshops on Intelligence and Security Informatics,
Springer, pp 377-383
[Wetzker08] Wetzker R, Plumbaum T, Korth A, Bauckhage C, Alpcan T, Metze F (2008a) Detecting trends in social
bookmarking systems using a probabilistic generative model and smoothing. In: Proceedings of 19th
International Conference on Pattern Recognition (ICPR 2008), IEEE, pp 1-4
[Hotho06] Hotho A, Jaschke R, Schmitz C, Stumme G (2006a) Information retrieval in folksonomies: Search and
ranking. In: Proceedings of the 3rd European Semantic Web Conference, Springer, Budva, Montenegro, LNCS,
vol 4011, pp 411-426
[Gill08] Gill, A.J., French R.M., Gergle, D., and Oberlander, J.: Identifying Emotional Characteristics from Short Blog
Texts. Proc. of the 30th Annual Conf. of the Cognitive Science Society, Washington DC (2008) 2237â2242
Editor's Notes
Starting from a k -clique, nodes are attached as long as they are reachable through clique adjacency (two k -cliques are adjacent if they share k -1nodes)
To sentiment analysis kai opinion mining ekfrazoun to idio. Me vasi to wikipedia Sentiment analysis or opinion mining refers to the application of natural language processing , computational linguistics , and text analytics to identify and extract subjective information in source materials. Apo dw kai katw krataw ton oro sentiment analysis
Ta relational networks anaparistoun to poia entities anaferontai mazi (nodes), to poso suxna anaferontai mazi (mikos akmwn), kai to poso prosfata xrhsimopoihthikan mazi (entasi tou edge)