Web Search Personalization via Social Bookmarking and Tagging Michael G. Noll

1
Web Search Personalization via Social
Bookmarking andTagging Michael G. Noll
Christoph MeinelHasso-Plattner-Institut an der
Universitat Potsdam, GermanyISWC 2007

• Advisor Prof. Hsin-Hsi Chen
• Reporter Yu-Hui Chang
• 2008/07/30

2
Introduction
3
Social bookmarking and tagging

• social bookmarking
• publicly sharing your bookmarks with others
• including any additional metadata
• tagging / folksonomies
• Users annotate Documents with with a flat,
  unstructured list of keywords called Tags

4
Web search personalization

• integration of user-specific data to improve
  results / advertising
• two main approaches
• modify user's query
• nyt gt new york times
• 2. re-rank search results based on user profile

5
Personalization Technique
6
Personalization

• Input
• user profile document profiles
• Via social bookmarking and tagging
• Algorithm
• calculate Similarity(user, document) for all docs
• sort documents by similarity from highest to
  lowest
• Output
• re-ranked search result list

7
Profile

• user profile
• Tagmarking
• A user search research internet security
• gthe/she can click single button
• to bookmark a document with auto translate tag
  research, internet, security
• document profiles
• Communicating with the bookmarking service over
  its web API

Tagmarking
8
Personalization

• Complete process of web search personalization
• every step done in real-time

9
Data Aggregation User Profile

• Users profile example

n documents
m tags
10
Data Aggregation Doc. Profile

• Documents profile example

n users
Mu
m tags
11
Similarity

• Similarity(u ,d)puT?pd
• pd simply normalization of document profile,
  reset matrix element with only 1 and 0 two values
  ( True or False)

12
Similarity example

• Similarity(u ,d)puT?pd

13 19 2 10 21 34
1 1 1 1 1 1
1301912010121034163
13
Similarity score properties

• the key factor unmodified user profile
• Promotes known and similar doc., demotes those
  unknown or non-similar doc.
• more sophisticated normalization for both user
  and document is on-going
• Score of unknown document gt 0!!!
• most critical factor in practice
• do we have sufficient data to make all this
  work?

14
Personalization

• system setup
• server social bookmarking service
• client browser add-on
• modification of search engine UI by updating the
  DOM tree of the search result pages in real-time

15
(No Transcript)
16
Re-rank example

• a user with a strong interest in information
  technology and network security

17
(No Transcript)
18
Experiment and Evaluation
19
Experiment

• Del.icio.us
• Public social bookmarking service
• Large user community
• Key question
• Quantitative analysis
• How many social annotations in practice?
• Qualitative analysis
• Quality evaluation

20
Quantitative analysis

• test set
• 140 popular tags on del.icio.us
• 1400 search results link (top 10 results )
• totaling
• 981,989 bookmarks
• 20,498 tag annotations (2,300 unique)

21
Quantitative analysis
22
Quantitative analysis
Percentage of links with at least 1 tag

• we can expect to personalize approx. 85 (in the
  1st page) per query in practice

23
Qualitative analysis

• For each query, participants were presented two
  search result lists
• original list from Google Search
• The personalized version
• 8 participants evaluate the top 10 results for
  13 queries each
• Participants job researchers, web masters,
  software developers, system administrators
• The average number of bookmarks for a participant
  was 153.

24
Qualitative analysis
Worse
Personalized version Better
Equal

• Some discussions
• Expert user profiles
• Disambiguate words and contexts

25
Conclusion
26
Conclusion

• proposed personalization approach is feasible and
  viable in practice
• already sufficient user-supplied metadata
  available
• initial evaluation of personalization quality
  shows very promising results
• Open Access
• http//www.michael-noll.com/dmoz100k06/ - data
  set
• http//www.michael-noll.com/delicious-api/ -
  scripts

27

• Thank you!!