Using subjectivity analysis to improve thread retrieval in online forums

Prakhar Biyani, Sumit Bhatia, Cornelia Caragea, Prasenjit Mitra

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Scopus citations

Abstract

Finding relevant threads in online forums is challenging for internet users due to a large number of threads discussing lexically similar topics but differing in the type of information they contain (e.g., opinions, facts, emotions). Search facilities need to take into account the match between users’ intent and the type of information contained in threads in addition to the lexical match between user queries and threads. We use intent match by incorporating subjectivity match between user queries and threads into a state-of-the-art forum thread retrieval model. Experimental results show that subjectivity match improves retrieval performance by over 10% as measured by different metrics.

Original languageEnglish (US)
Title of host publicationAdvances in Information Retrieval - 37th European Conference on IR Research, ECIR 2015, Proceedings
EditorsAllan Hanbury, Andreas Rauber, Gabriella Kazai, Norbert Fuhr
PublisherSpringer Verlag
Pages495-500
Number of pages6
ISBN (Electronic)9783319163536
DOIs
StatePublished - 2015
Event37th European Conference on Information Retrieval Research, ECIR 2015 - Vienna, Austria
Duration: Mar 29 2015Apr 2 2015

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9022
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other37th European Conference on Information Retrieval Research, ECIR 2015
Country/TerritoryAustria
CityVienna,
Period3/29/154/2/15

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Using subjectivity analysis to improve thread retrieval in online forums'. Together they form a unique fingerprint.

Cite this