biased lexrank : passage retrieval using random walks with question-based priors

16
Intelligent Database Systems Presenter : JHOU, YU-LIANG Authors : Jahna Otterbacher a , Gunes Erkan b , Dragomir R. Radev 2009, IPM Biased LexRank: Passage Retrieval using Random Walks with Question-Based Priors

Upload: tariq

Post on 08-Jan-2016

59 views

Category:

Documents


0 download

DESCRIPTION

Biased LexRank : Passage Retrieval using Random Walks with Question-Based Priors. Presenter : JHOU, YU-LIANG Authors : Jahna Otterbacher a , Gunes Erkan b , Dragomir R. Radev 2009, IPM. Outlines. Motivation Objectives Methodology Experimental Result Conclusions Comments. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Presenter : JHOU, YU-LIANG

Authors : Jahna Otterbacher a , Gunes Erkan b , Dragomir R. Radev

2009, IPM

Biased LexRank: Passage Retrieval usingRandom Walks with Question-Based Priors

Page 2: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Outlines

MotivationObjectivesMethodology Experimental ResultConclusionsComments

Page 3: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Motivation• Text summarization is one of the hardest problems in

information retrieval, because it is not very well-defined.

• There are various definitions of text summarization resulting from different approaches to solving the problem.

• There is often no agreement as to what a good summary is even when we are dealing with a particular definition of the problem.

Page 4: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

ObjectivesUsing biased LexRank on achieving text summarization and retrieval QA more effect .

Page 5: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

LexRank

Page 6: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Biased LexRank

Page 7: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Biased LexRank-application-QA

QA systems is to retrieve the sentences that potentially contain the answer to the question .

Page 8: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Passage retrieval-summarization

Computing link weights

Page 9: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Passage retrieval- Question answering

Page 10: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Experimental-result

biased LexRank v.s human summarizers

Page 11: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Experimental-QAcorpus

Page 12: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Experimental Result effect of similarity for QA

Page 13: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Experimental Result LexRank Versus the Baseline Approach

Page 14: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Experimental ResultLexRank Versus the Baseline Approach

Page 15: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

Conclusions

In the paper, we have also demonstrated the

effectiveness of our method as applied to two classical

IR problems, extractive text summarization and passage

retrieval for question answering.

Page 16: Biased  LexRank : Passage Retrieval using Random Walks with Question-Based Priors

Intelligent Database Systems Lab

CommentsI think the method improved retrieval performance and comparable to human summarizers.Applications

- Text summarization- Information retrieval