Analysis and Monetization of Social Data

1,420 views

Published on

Amit Sheth, "Analysis and Monetization of Social Data" , Panel on Semantifying Social Network, 2009 Semantic Technology Conference, San Jose, CA, June 22, 2009.
http://www.semantic-conference.com/ataglance/

Published in: Technology, Business
0 Comments
1 Like
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
1,420
On SlideShare
0
From Embeds
0
Number of Embeds
13
Actions
Shares
0
Downloads
48
Comments
0
Likes
1
Embeds 0
No embeds

No notes for slide
  • monetizing social data, advertising on social networks, spatio-temporal-thematic analysis of social data, Twitris, analysis of user generated content
  • Social networks are growing at a rabid pace. Creating lots of data: observational data, people – people relationship data, user generated content etc
  • Using profile information - Micro targeted ads (demographic-based) -- Sponsored ads What we are seeing today: Is not apparently appealing to its members – show / tell stats that is showing that no one is paying atn to ads here What is happening here -- No intents
  • A concert – show tickets; food – red robin ; intent to purchase – body shop
  • How people write? Why they write and what they write? Only content based; there are other techniques such as link based analysis, social behavior influence
  • We understand there are other dimensions to this analysis – identifying objectionable content and unfavorable sentiments .. Here we will focus on these two problems
  • Information seeking Intent is to solicit responses concerning a question that addresses the information needs of the user. The query can be one asking for information toward the end goal of comparisons, transactions, locating a web page etc. Information sharing Intent is to inform. Users are typically sharing information or opinion about a product, an experience, promotions etc. Transactional Intent is to express an explicit buy, sell or trade intents. The goal is to seek responses that will provide cues leading to an offline (outside-network) transaction.
  • Find intent, remove off topic noise
  • Follow this with two snapshots showing temporal and spatial variations
  • What is new and interesting? What’s a region paying attention to today? What are people most excited or concerned about? Why an entity’s perception changing over time in any region?
  • Analysis and Monetization of Social Data

    1. 1. Analysis and Monetization of Social Data Panel on Semantifying Social Networks, 2009 Semantic Technology Conference , San Jose CA, June 22, 2009 <ul><li>Amit P. Sheth </li></ul><ul><li>Lexis-Nexis Ohio Eminent Scholar </li></ul><ul><li>Director, Kno.e.sis Center, Wright State University </li></ul><ul><li>Thanks: Meena Nagarajan </li></ul>
    2. 2. 222 MILLION FACEBOOK USERS 4000000 twitter users 3 Million tweets a day 52,000 F8 APPLICATIONS AND COUNTING
    3. 4. Intents in User Activity Elsewhere June 01, 2009
    4. 5. What why and how people write <ul><li>Cultural Entities </li></ul><ul><li>Word Usages in self-presentation </li></ul><ul><li>Slang sentiments </li></ul><ul><li>Intentions </li></ul>
    5. 6. Work and Preliminary Results in… <ul><li>Identifying intents behind user posts on social networks </li></ul><ul><ul><li>Pull UGC with most monetization potential </li></ul></ul><ul><li>Identifying keywords for advertizing in user-generated content </li></ul><ul><ul><li>Interpersonal communication & off-topic chatter </li></ul></ul>
    6. 7. Identifying Monetizable Intents <ul><li>Scribe Intent not same as Web Search Intent 1 </li></ul><ul><li>People write sentences, not keywords or phrases </li></ul><ul><li>Presence of a keyword does not imply navigational / transactional intents </li></ul><ul><ul><li>‘ am thinking of getting X’ ( transactional ) </li></ul></ul><ul><ul><li>‘ i like my new X’ (information sharing) </li></ul></ul><ul><ul><li>‘ what do you think about X’ ( information seeking ) </li></ul></ul>1 B. 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.
    7. 8. From X to Action Patterns <ul><li>Action patterns surrounding an entity </li></ul><ul><li>How questions are asked and not topic words that indicate what the question is about </li></ul><ul><li>“ where can I find a chotto psp cam” </li></ul><ul><ul><li>User post also has an entity </li></ul></ul>
    8. 9. Off topic noise – topical keywords <ul><li>Google AdSense ads for user post vs. extracted topical keywords </li></ul>
    9. 10. 8X Generated Interest <ul><li>Using profile ads </li></ul><ul><ul><li>Total of 56 ad impressions </li></ul></ul><ul><ul><li>7% of ads generated interest </li></ul></ul><ul><li>Using authored posts </li></ul><ul><ul><li>Total of 56 ad impressions </li></ul></ul><ul><ul><li>43% of ads generated interest </li></ul></ul><ul><li>Using topical keywords from authored posts </li></ul><ul><ul><li>Total of 59 ad impressions </li></ul></ul><ul><ul><li>59% of ads generated interest </li></ul></ul>
    10. 11. <ul><li>and then there is </li></ul><ul><li>space (where) </li></ul><ul><li>time (when) </li></ul><ul><ul><li>theme (what) </li></ul></ul>
    11. 13. <ul><li>twitris: spatio-temporal integration of twitter data “surrounding” an event </li></ul><ul><li>http://twitris.dooduh.com </li></ul>
    12. 14. Studying social signals <ul><li>What is new and interesting? </li></ul><ul><li>What’s a region paying attention to today? What are people most excited or concerned about? </li></ul><ul><li>Why an entity’s perception changing over time in any region? </li></ul>
    13. 15. Image Metadata latitude: 18° 54′ 59.46″ N, longitude: 72° 49′ 39.65″ E Geocoder (Reverse Geo-coding) Address to location database 18 Hormusji Street, Colaba Nariman House Identify and extract information from tweets Spatio-Temporal Analysis Structured Meta Extraction Income Tax Office Vasant Vihar
    14. 17. <ul><li>domain models to enhance thematic </li></ul><ul><li>relationships </li></ul>
    15. 18. <ul><li>who creates? </li></ul>
    16. 19. <ul><li>I will, you will, WE will </li></ul>
    17. 20. More at library@Kno.e.sis: http://knoesis.org <ul><li>A. Sheth, &quot;A Playground for Mobile Sensors, Human Computing, and Semantic Analytics&quot;, IEEE Internet Computing, July/August 2009, pp. 80-85. </li></ul><ul><li>M. Nagarajan, K. Baid, A. P. Sheth, and S. Wang, &quot;Monetizing User Activity on Social Networks - Challenges and Experiences“, 2009 IEEE/WIC/ACM International Conference on Web Intelligence WI-09 , Milan, Italy </li></ul><ul><li>M. Nagarajan, et al. Spatio-Temporal-Thematic Analysis of Citizen-Sensor Data - Challenges and Experiences, Web Information Systems Engineering- WISE-2009 , Poznan, Poland (to appear). </li></ul>http://knoesis.org/research/semweb/projects/socialmedia/

    ×