4. +Introduction
n Faceted search helps users by offering
drill-down options as a complement to
the keyword input box.
4
5. +Introduction
n However, this idea is not well explored
for general web search.
n heterogeneous nature
5
6. +Introduction
n However, this idea is not well explored
for general web search.
n heterogeneous nature
6
baggage allowance
所有航線
所有航線
國際航線
國內航線
貨運公司
行李類型
7. +Introduction
n However, this idea is not well explored
for general web search.
n heterogeneous nature
7
baggage allowance
所有航線
所有航線
國際航線
國內航線
貨運公司
行李類型
← query
← facet
← facet term
↓ search result ( ducument)
8. +Introduction
n Goal :
n query-dependent automatic facet generation
n user feedback on these query facets into
document ranking
8
13. +
Facet
Generation
Facet
Feedback
Evaluation
13
n Step 1 : Extracting Candidates
n applied both textual and HTML patterns on
the top search results
14. +
Facet
Generation
Facet
Feedback
Evaluation
14
n Step 1 : Extracting Candidates
n query : “mars landing”
n search results
n “ Mars rovers such as Curiosity, Opportunity
and Spirit ”
n candidate facets
n C : { Curiosity, Opportunity, Spirit }
15. +
Facet
Generation
Facet
Feedback
Evaluation
15
n Step 1 : Extracting Candidates
n the candidate query facets extracted.
n noisy
n non-relevant to the issued query
n terms be not members of the same class
23. +
Facet
Generation
Facet
Feedback
Evaluation
23
n Input : Document, Query, User Selection
n Document = one of search result
n Boolean Filtering Model
n Soft Ranking Model
n Output : the score of each document
24. +
Facet
Generation
Facet
Feedback
Evaluation
24
n Boolean Filtering Model
n Fu denotes the set of feedback facets which
user selected
n condition B can be either AND, OR, or A+O
n S(D, Q) is the score returned by the original
retrieval model
25. +
Facet
Generation
Facet
Feedback
Evaluation
25
n Soft Ranking Model
n λ is a parameter for adjusting the weight
n SE(D, Fu) is the expansion part which captures
the relevance between the document and
feedback facet
26. +
Facet
Generation
Facet
Feedback
Evaluation
26
n Input : Documents, Query, User Selection
n Boolean Filtering Model
n Soft Ranking Model
n Output : the score of each document
27. +
Facet
Generation
Facet
Feedback
Evaluation
27
n Intrinsic Evaluation
n Ground Truth: query facets are constructed
by human annotators
n annotators are asked to group or re-group
terms in the pool into preferred query facets.
n pooling facets generated by the different systems
n compared with facets generated by different
systems
28. +
Facet
Generation
Facet
Feedback
Evaluation
28
n Extrinsic Evaluation
n User Model
n The user model describes how a user selects
feedback terms from facets, based on which we can
estimate the time cost for the user.
↑
time for selecting terms
time for scanning facet
↓
29. +
Facet
Generation
Facet
Feedback
Evaluation
29
n Extrinsic Evaluation
n Oracle Feedback and Annotator Feedback
n Oracle feedback model only selected effective terms
as feedback.
n The annotator is asked to select all the terms from
the facets that would help address the information
need.
33. +Facet Generation Models
33
based on annotator feedback
and SF feedback model
based on oracle feedback
and SF feedback model.
34. +Facet Generation Models
34
Our experiments testify to the
potential of Faceted Web Search.
based on annotator feedback
and SF feedback model
based on oracle feedback
and SF feedback model.
38. +Conclusion
n This paper proposed Faceted Web
Search.
n an extension of faceted search to the general
Web
n query-dependent automatic facet
generation
n feedback on these query facets into
document ranking
38