Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Collecting Community Wisdom: Integrating Social Search & Social Navigation Jill Freyne, Rosta Farzan, Peter Brusilovsky, B...
Motivations <ul><li>Information Overload Problem </li></ul><ul><li>Potential in harnessing user activity patterns to drive...
Social Browsing <ul><li>Form of Social Navigation </li></ul><ul><li>Visualizes usage paths in a digital environment </li><...
KnowlegeSea II <ul><li>Footprint based support </li></ul><ul><ul><li>Monitors students as they browse the collection </li>...
Social Search <ul><li>Collaborative Web search </li></ul><ul><li>Communities of like minded users searching together </li>...
I-SPY <ul><li>Monitors users as they search </li></ul><ul><li>Post Processing Engine </li></ul><ul><li>Re-ranks result lis...
ACM DL <ul><li>ACM DL -  a vast collection of citations and full text from ACM journal and newsletter articles and confere...
Searching
Browsing
Integrated Social Information Access <ul><li>“ Community Wisdom” collected by the search component should be used to suppo...
Components <ul><li>Search Component </li></ul><ul><ul><li>I-SPY Search technology </li></ul></ul><ul><ul><li>Re-rank CACM ...
Search Component Architecture CWS ENGINE CACM  SEARCH ENGINE q q` R M R H R T Hit Matrix q
Browse Component Architecture KNOWLEDGE SEA ENGINE Navigation  (Browsing & Annotation) Records
Integration - Search CACM Search Engine Navigation Engine CWS  Engine Hit Matrix Navigation  (Browsing & Annotation) Recor...
Integration - Browsing Users start with browsing Navigation Engine CWS Engine Search Hit Matrix Navigation  (Browsing & An...
Annotations
Search Support Relevance:  100% Related Queries: Social Navigation Computational Wear Last Selection:  30 mins ago Last Br...
Browsing Support Queries: personalization business customer adaptive web Browse Popularity : 5% Annotation: …………
Browse – Search Cycle
Evaluation <ul><li>Subjects:  30  students enrolled in “Introduction to Multimedia” course </li></ul><ul><li>Task:  Produc...
Review Relevance <ul><li>54 articles returned </li></ul><ul><li>Each examined and assigned relevance score </li></ul><ul><...
Search and Browsing Effort <ul><li>Users in the control group had to do more work to return (less relevant) results. </li>...
So? <ul><li>By adding social support to the CACM DL users found relevant information faster and with less user effort </li...
Subjective Data Analysis <ul><li>Questionnaire </li></ul><ul><ul><li>Designed to capture user opinion on the system </li><...
Searching
Browsing
Conclusions <ul><li>Different community-based information access technologies can be used in an integrated system to reinf...
Future work <ul><li>Larger user study </li></ul><ul><ul><li>Larger, longer term study </li></ul></ul><ul><li>Domain diverg...
Collecting Community Wisdom: Integrating Social Search & Social Navigation Jill Freyne,  Rosta Farzan, Peter Brusilovsky, ...
User Feedback <ul><li>All subjects noticed the social icons </li></ul><ul><li>All subjects often/always drawn to icons </l...
Social Search <ul><li>Monitors users as they search, recording their queries and result selection pairs </li></ul><ul><li>...
Search and Browsing Efforts <ul><li>58% of users did not look beyond result page 1 </li></ul><ul><li>88% did not look beyo...
 
Upcoming SlideShare
Loading in …5
×

Jill Freyne - Collecting community wisdom: integrating social search and social navigation

3,054 views

Published on

WebCamp presentation on an augmented search aided by previous searches by social connections.

Published in: Technology, Education
  • Dating direct: ❶❶❶ http://bit.ly/39pMlLF ❶❶❶
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • Dating for everyone is here: ❶❶❶ http://bit.ly/39pMlLF ❶❶❶
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here

Jill Freyne - Collecting community wisdom: integrating social search and social navigation

  1. 1. Collecting Community Wisdom: Integrating Social Search & Social Navigation Jill Freyne, Rosta Farzan, Peter Brusilovsky, Barry Smyth & Maurice Coyle University College Dublin & University of Pittsburgh
  2. 2. Motivations <ul><li>Information Overload Problem </li></ul><ul><li>Potential in harnessing user activity patterns to drive social information access tools </li></ul><ul><li>Independence of social systems </li></ul><ul><ul><li>Eurekster, Del.icio.us, Shadows, I-SPY, MySpace, ebay,….. </li></ul></ul>
  3. 3. Social Browsing <ul><li>Form of Social Navigation </li></ul><ul><li>Visualizes usage paths in a digital environment </li></ul><ul><li>KnowlegeSea II Brusilovsky et al. AH 2004 </li></ul><ul><ul><li>Information access system for class mates providing access to open corpus textbooks </li></ul></ul><ul><ul><li>Supports information access through visualization, search, annotation and browsing </li></ul></ul>
  4. 4. KnowlegeSea II <ul><li>Footprint based support </li></ul><ul><ul><li>Monitors students as they browse the collection </li></ul></ul><ul><li>Annotation based support </li></ul><ul><ul><li>Allows students to leave specific traces on parts of the system </li></ul></ul>
  5. 5. Social Search <ul><li>Collaborative Web search </li></ul><ul><li>Communities of like minded users searching together </li></ul><ul><li>I-SPY Smyth et al. UMUAI 2004 </li></ul><ul><ul><li>Social search engine where communities of searchers can benefit from each others wisdom in terms of search queries and result selections </li></ul></ul>
  6. 6. I-SPY <ul><li>Monitors users as they search </li></ul><ul><li>Post Processing Engine </li></ul><ul><li>Re-ranks result lists from underlying search engines to reflect community preferences </li></ul>
  7. 7. ACM DL <ul><li>ACM DL - a vast collection of citations and full text from ACM journal and newsletter articles and conference proceedings . </li></ul><ul><li>2 access strategies </li></ul><ul><ul><li>Browsing – through archive lists, tables of contents </li></ul></ul><ul><ul><li>Searching – basic and advanced search features </li></ul></ul>
  8. 8. Searching
  9. 9. Browsing
  10. 10. Integrated Social Information Access <ul><li>“ Community Wisdom” collected by the search component should be used to support navigation in the browsing component and vice versa </li></ul><ul><li>Social Search and Social Browsing should be seamlessly integrated at the interface level </li></ul><ul><li>Social Support in the form of icons/cues </li></ul>
  11. 11. Components <ul><li>Search Component </li></ul><ul><ul><li>I-SPY Search technology </li></ul></ul><ul><ul><li>Re-rank CACM result-list and add search icons </li></ul></ul><ul><li>Browsing Component </li></ul><ul><ul><li>Knowledge Sea II social navigation technology </li></ul></ul><ul><ul><li>Provide footprint and annotation support and add browse icons </li></ul></ul>
  12. 12. Search Component Architecture CWS ENGINE CACM SEARCH ENGINE q q` R M R H R T Hit Matrix q
  13. 13. Browse Component Architecture KNOWLEDGE SEA ENGINE Navigation (Browsing & Annotation) Records
  14. 14. Integration - Search CACM Search Engine Navigation Engine CWS Engine Hit Matrix Navigation (Browsing & Annotation) Records Result Set Re-ranked result With social search icons Re-ranked result With social search and social navigation icons Users start with search
  15. 15. Integration - Browsing Users start with browsing Navigation Engine CWS Engine Search Hit Matrix Navigation (Browsing & Annotation) Records Adding social search icons Adding social navigation icons Back to search Query based navigation
  16. 16. Annotations
  17. 17. Search Support Relevance: 100% Related Queries: Social Navigation Computational Wear Last Selection: 30 mins ago Last Browse: 1 day ago Last Annotation: 1 day ago Browse Popularity 30% Annotations: Relevant to previous work by Dourish and Chalmers in Social Navigation
  18. 18. Browsing Support Queries: personalization business customer adaptive web Browse Popularity : 5% Annotation: …………
  19. 19. Browse – Search Cycle
  20. 20. Evaluation <ul><li>Subjects: 30 students enrolled in “Introduction to Multimedia” course </li></ul><ul><li>Task: Produce a literature review on the topic of “the social web” in 1 hour using the ACM social site </li></ul><ul><li>Groups: </li></ul><ul><ul><li>Control group - no social support (15 users) </li></ul></ul><ul><ul><li>Experimental Group - all social support enabled (15 users) </li></ul></ul>
  21. 21. Review Relevance <ul><li>54 articles returned </li></ul><ul><li>Each examined and assigned relevance score </li></ul><ul><li>22.4% relative increase in very relevant papers </li></ul>
  22. 22. Search and Browsing Effort <ul><li>Users in the control group had to do more work to return (less relevant) results. </li></ul>1.26 1.56 Ave selected result page 7.0 12.53 # links browsed 2.37 2.33 Query length 9.13 13.2 # queries Exp Ctrl
  23. 23. So? <ul><li>By adding social support to the CACM DL users found relevant information faster and with less user effort </li></ul>
  24. 24. Subjective Data Analysis <ul><li>Questionnaire </li></ul><ul><ul><li>Designed to capture user opinion on the system </li></ul></ul><ul><li>70% of the exp group found social ACM system useful for task </li></ul><ul><li>40% of ctrl group reported the same </li></ul>
  25. 25. Searching
  26. 26. Browsing
  27. 27. Conclusions <ul><li>Different community-based information access technologies can be used in an integrated system to reinforce each other and provide unique added-value to the users </li></ul><ul><li>Integration of both technologies was achieved through seamless connection from search to browsing and browsing to search </li></ul>
  28. 28. Future work <ul><li>Larger user study </li></ul><ul><ul><li>Larger, longer term study </li></ul></ul><ul><li>Domain divergence </li></ul><ul><ul><li>Introduce social support to other domains </li></ul></ul><ul><ul><ul><li>Multimedia, other academic domains </li></ul></ul></ul>
  29. 29. Collecting Community Wisdom: Integrating Social Search & Social Navigation Jill Freyne, Rosta Farzan, Peter Brusilovsky, Barry Smyth & Maurice Coyle University College Dublin & University of Pittsburgh
  30. 30. User Feedback <ul><li>All subjects noticed the social icons </li></ul><ul><li>All subjects often/always drawn to icons </li></ul><ul><li>Users reported that they found it easier to locate information on the ACM DL site with the icons than without. </li></ul>
  31. 31. Social Search <ul><li>Monitors users as they search, recording their queries and result selection pairs </li></ul><ul><li>Uses past search interaction to re-rank the results of an underlying search engine to reflect the preferences of a community of searchers </li></ul>
  32. 32. Search and Browsing Efforts <ul><li>58% of users did not look beyond result page 1 </li></ul><ul><li>88% did not look beyond result page 3 </li></ul>

×