sigir05 if

Information about sigir05 if

Published on November 21, 2007

Author: Brainy007

Source: authorstream.com

Content

Context-Sensitive IR using Implicit Feedback :  Context-Sensitive IR using Implicit Feedback Xuehua Shen, Bin Tan, ChengXiang Zhai Department of Computer Science University of Illinois, Urbana-Champaign Problem of Context-Independent Search:  Problem of Context-Independent Search Jaguar Put Search in Context:  Other Context Info: Dwelling time Mouse movement Put Search in Context Apple software Hobby … Problem Definition:  Problem Definition … How to model and use all the information? e.g., Apple software e.g., Apple - Mac OS X The Apple Mac OS X product page. Describes features in the current version of Mac OS X, a screenshot gallery, latest software downloads, and a directory of ... Outline:  Outline Four contextual statistical language models Experiment design and results Summary and future work Retrieval Model:  Retrieval Model Qk D θQk θD Similarity Measure Results Basis: Unigram language model + KL divergence Fixed Coefficient Interpolation (FixInt):  Fixed Coefficient Interpolation (FixInt) Qk Bayesian Interpolation (BayesInt):  Bayesian Interpolation (BayesInt) Intuition: if the current query Qk is longer, we should trust Qk more Online Bayesian Update (OnlineUp):  Online Bayesian Update (OnlineUp) Intuition: continuous belief update about user information need Batch Bayesian Update (BatchUp):  Batch Bayesian Update (BatchUp) C1 C2 Intuition: clickthrough data may not decay Data Set of Evaluation:  Data Set of Evaluation Data collection: TREC AP88-90 Topics: 30 hard topics of TREC topics 1-150 System: search engine + RDBMS Context: Query and clickthrough history of 3 participants. Experiment Design:  Experiment Design Models: FixInt, BayesInt, OnlineUp and BatchUp Performance Comparison: Qk vs. Qk+HQ+HC Evaluation Metrics: MAP and [email protected] docs Overall Effect of Search Context:  Overall Effect of Search Context Interaction history helps system improve retrieval accuracy BayesInt better than FixInt; BatchUp better than OnlineUp Using Clickthrough Data Only:  Using Clickthrough Data Only BayesInt (=0.0,=5.0) Sensitivity of BatchUp Parameters:  Sensitivity of BatchUp Parameters BatchUp is stable with different parameter settings Best performance is achieved when =2.0; =15.0 Summary:  Summary Propose four contextual language models to exploit user interaction history for contextual search Construct an evaluation dataset based on TREC data (http://sifaka.cs.uiuc.edu/ir/ucair/QCHistory.zip) Experiment results show that user interaction history, especially clickthrough data, can improve the retrieval accuracy Future Work:  Future Work Study a general framework for interactive information retrieval Study more sophisticated models to incorporate context information Build a system on the client side to capture and exploit user context information Slide18:  Thank you ! The End

Related presentations


Other presentations created by Brainy007

impact irrigation
23. 08. 2007
0 views

impact irrigation

Chapter 10
06. 11. 2007
0 views

Chapter 10

Biomes Notes
10. 10. 2007
0 views

Biomes Notes

national courier
03. 10. 2007
0 views

national courier

Context aware Event Notification
05. 10. 2007
0 views

Context aware Event Notification

Tilde overview
08. 10. 2007
0 views

Tilde overview

India Little Stories Big Dreams
12. 10. 2007
0 views

India Little Stories Big Dreams

Trainer of Training Admin
04. 09. 2007
0 views

Trainer of Training Admin

PresentaciÃn Chile ppp
22. 10. 2007
0 views

PresentaciÃn Chile ppp

munoz
01. 12. 2007
0 views

munoz

Triage mobisys
29. 10. 2007
0 views

Triage mobisys

plead
29. 10. 2007
0 views

plead

life in tudor times
23. 08. 2007
0 views

life in tudor times

Greek Mythology mlm
23. 08. 2007
0 views

Greek Mythology mlm

em ipr gm indigenousseed
17. 10. 2007
0 views

em ipr gm indigenousseed

seaman 1
07. 11. 2007
0 views

seaman 1

ciaa summary en
04. 12. 2007
0 views

ciaa summary en

mccarthywipo
27. 11. 2007
0 views

mccarthywipo

chatguide KS3
28. 12. 2007
0 views

chatguide KS3

WB20070713
25. 10. 2007
0 views

WB20070713

0903BROWNFIELDS8DALE
01. 01. 2008
0 views

0903BROWNFIELDS8DALE

Black History Month slides
02. 01. 2008
0 views

Black History Month slides

Ch17 Acid Base A
03. 01. 2008
0 views

Ch17 Acid Base A

birds mammals 2007
03. 01. 2008
0 views

birds mammals 2007

The Nineteenth Century
07. 01. 2008
0 views

The Nineteenth Century

lhj Tudor Fashion
23. 08. 2007
0 views

lhj Tudor Fashion

sigirtutorial2004
23. 08. 2007
0 views

sigirtutorial2004

436group6
03. 08. 2007
0 views

436group6

071F1301 Fire Commands
07. 11. 2007
0 views

071F1301 Fire Commands

conf faking ellingson
03. 08. 2007
0 views

conf faking ellingson

talk
16. 10. 2007
0 views

talk

Caldwell1
15. 10. 2007
0 views

Caldwell1

H113k
07. 12. 2007
0 views

H113k

Dangerous MarketsMcKinseyl
15. 10. 2007
0 views

Dangerous MarketsMcKinseyl

202636
24. 02. 2008
0 views

202636

976 Tues Costello
28. 02. 2008
0 views

976 Tues Costello

Llewellyn
07. 11. 2007
0 views

Llewellyn

TRL
06. 03. 2008
0 views

TRL

midterm review 1
02. 11. 2007
0 views

midterm review 1

ProducingGender
19. 02. 2008
0 views

ProducingGender

politru
20. 03. 2008
0 views

politru

HauWei 110306
26. 10. 2007
0 views

HauWei 110306

mail defence methods
07. 10. 2007
0 views

mail defence methods

mishkin ppt09
09. 04. 2008
0 views

mishkin ppt09

chinarisk
10. 04. 2008
0 views

chinarisk

rt1 maeda
13. 04. 2008
0 views

rt1 maeda

mabrouk undergraduate research
30. 10. 2007
0 views

mabrouk undergraduate research

8 Stock Markets
18. 04. 2008
0 views

8 Stock Markets

Genmeet 2005 Marques
22. 04. 2008
0 views

Genmeet 2005 Marques

PaulJourdan
28. 04. 2008
0 views

PaulJourdan

afghan
27. 02. 2008
0 views

afghan

ai intro
30. 04. 2008
0 views

ai intro

Munich Airport
16. 10. 2007
0 views

Munich Airport

tudor banquet costume
23. 08. 2007
0 views

tudor banquet costume

New Arrivals
02. 10. 2007
0 views

New Arrivals

dansereau
18. 06. 2007
0 views

dansereau

brantley 2
18. 06. 2007
0 views

brantley 2

anthrax
18. 06. 2007
0 views

anthrax

Anne geneve cours6
18. 06. 2007
0 views

Anne geneve cours6

anne geneve cours4
18. 06. 2007
0 views

anne geneve cours4

amm pres wilson apnic rpt
18. 06. 2007
0 views

amm pres wilson apnic rpt

91 Presentation Desc
18. 06. 2007
0 views

91 Presentation Desc

WCU07Recordkeeping
16. 02. 2008
0 views

WCU07Recordkeeping

DC1 Talk
18. 06. 2007
0 views

DC1 Talk

Learn Geometric Patterns
15. 06. 2007
0 views

Learn Geometric Patterns

Learn Fractions
15. 06. 2007
0 views

Learn Fractions

Learn Fractions XIV
15. 06. 2007
0 views

Learn Fractions XIV

Learning Percent II
15. 06. 2007
0 views

Learning Percent II

Fraction X
15. 06. 2007
0 views

Fraction X

foresbiome
15. 06. 2007
0 views

foresbiome

favorite animal
15. 06. 2007
0 views

favorite animal

Fact Families
15. 06. 2007
0 views

Fact Families

Asia Barabash
26. 10. 2007
0 views

Asia Barabash

ContentFile57
16. 10. 2007
0 views

ContentFile57

Learning Geological Features
15. 06. 2007
0 views

Learning Geological Features

Learn Fractions VIII
15. 06. 2007
0 views

Learn Fractions VIII

Foldable for Quadrilaterals
15. 06. 2007
0 views

Foldable for Quadrilaterals

lecture6
15. 11. 2007
0 views

lecture6

ASPLOS 2000
18. 06. 2007
0 views

ASPLOS 2000

Dude watercontrol team
30. 12. 2007
0 views

Dude watercontrol team

cuc kl 0105 kka
04. 09. 2007
0 views

cuc kl 0105 kka

tudor monarchs afterhenry
23. 08. 2007
0 views

tudor monarchs afterhenry

McKee 10 11 05
30. 10. 2007
0 views

McKee 10 11 05

WN NLTN March 9 2004
21. 10. 2007
0 views

WN NLTN March 9 2004

sbdc naples june05
22. 10. 2007
0 views

sbdc naples june05

PARCCppt040506
29. 10. 2007
0 views

PARCCppt040506