Unsupervised Solution Post Identification from Discussion Forums

Deepak Padmanabhan, Karthik Visweswariah

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

6 Citations (Scopus)

Abstract

Discussion forums have evolved into a dependablesource of knowledge to solvecommon problems. However, only a minorityof the posts in discussion forumsare solution posts. Identifying solutionposts from discussion forums, hence, is animportant research problem. In this paper,we present a technique for unsupervisedsolution post identification leveraginga so far unexplored textual feature, thatof lexical correlations between problemsand solutions. We use translation modelsand language models to exploit lexicalcorrelations and solution post characterrespectively. Our technique is designedto not rely much on structural featuressuch as post metadata since suchfeatures are often not uniformly availableacross forums. Our clustering-based iterativesolution identification approach basedon the EM-formulation performs favorablyin an empirical evaluation, beatingthe only unsupervised solution identificationtechnique from literature by a verylarge margin. We also show that our unsupervisedtechnique is competitive againstmethods that require supervision, outperformingone such technique comfortably.
LanguageEnglish
Title of host publicationProceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
PublisherAssociation for Computational Linguistics
Pages155-164
Number of pages10
Publication statusPublished - 2014
Event52nd Annul Meeting of the Association of Computational Linguistics 2014 - Maryland, Baltimore, United States
Duration: 22 Jun 2014 → 27 Jun 2014

Conference

Conference52nd Annul Meeting of the Association of Computational Linguistics 2014
CountryUnited States
CityBaltimore
Period22/06/2014 → 27/06/2014

Fingerprint

Metadata

Cite this

Padmanabhan, D., & Visweswariah, K. (2014). Unsupervised Solution Post Identification from Discussion Forums. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 155-164). Association for Computational Linguistics.
Padmanabhan, Deepak ; Visweswariah, Karthik. / Unsupervised Solution Post Identification from Discussion Forums. Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Computational Linguistics, 2014. pp. 155-164
@inproceedings{4b381dd9bac846cdaf874cc1f5a3a44a,
title = "Unsupervised Solution Post Identification from Discussion Forums",
abstract = "Discussion forums have evolved into a dependablesource of knowledge to solvecommon problems. However, only a minorityof the posts in discussion forumsare solution posts. Identifying solutionposts from discussion forums, hence, is animportant research problem. In this paper,we present a technique for unsupervisedsolution post identification leveraginga so far unexplored textual feature, thatof lexical correlations between problemsand solutions. We use translation modelsand language models to exploit lexicalcorrelations and solution post characterrespectively. Our technique is designedto not rely much on structural featuressuch as post metadata since suchfeatures are often not uniformly availableacross forums. Our clustering-based iterativesolution identification approach basedon the EM-formulation performs favorablyin an empirical evaluation, beatingthe only unsupervised solution identificationtechnique from literature by a verylarge margin. We also show that our unsupervisedtechnique is competitive againstmethods that require supervision, outperformingone such technique comfortably.",
author = "Deepak Padmanabhan and Karthik Visweswariah",
year = "2014",
language = "English",
pages = "155--164",
booktitle = "Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
publisher = "Association for Computational Linguistics",

}

Padmanabhan, D & Visweswariah, K 2014, Unsupervised Solution Post Identification from Discussion Forums. in Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Computational Linguistics, pp. 155-164, 52nd Annul Meeting of the Association of Computational Linguistics 2014, Baltimore, United States, 22/06/2014.

Unsupervised Solution Post Identification from Discussion Forums. / Padmanabhan, Deepak; Visweswariah, Karthik.

Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Computational Linguistics, 2014. p. 155-164.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Unsupervised Solution Post Identification from Discussion Forums

AU - Padmanabhan, Deepak

AU - Visweswariah, Karthik

PY - 2014

Y1 - 2014

N2 - Discussion forums have evolved into a dependablesource of knowledge to solvecommon problems. However, only a minorityof the posts in discussion forumsare solution posts. Identifying solutionposts from discussion forums, hence, is animportant research problem. In this paper,we present a technique for unsupervisedsolution post identification leveraginga so far unexplored textual feature, thatof lexical correlations between problemsand solutions. We use translation modelsand language models to exploit lexicalcorrelations and solution post characterrespectively. Our technique is designedto not rely much on structural featuressuch as post metadata since suchfeatures are often not uniformly availableacross forums. Our clustering-based iterativesolution identification approach basedon the EM-formulation performs favorablyin an empirical evaluation, beatingthe only unsupervised solution identificationtechnique from literature by a verylarge margin. We also show that our unsupervisedtechnique is competitive againstmethods that require supervision, outperformingone such technique comfortably.

AB - Discussion forums have evolved into a dependablesource of knowledge to solvecommon problems. However, only a minorityof the posts in discussion forumsare solution posts. Identifying solutionposts from discussion forums, hence, is animportant research problem. In this paper,we present a technique for unsupervisedsolution post identification leveraginga so far unexplored textual feature, thatof lexical correlations between problemsand solutions. We use translation modelsand language models to exploit lexicalcorrelations and solution post characterrespectively. Our technique is designedto not rely much on structural featuressuch as post metadata since suchfeatures are often not uniformly availableacross forums. Our clustering-based iterativesolution identification approach basedon the EM-formulation performs favorablyin an empirical evaluation, beatingthe only unsupervised solution identificationtechnique from literature by a verylarge margin. We also show that our unsupervisedtechnique is competitive againstmethods that require supervision, outperformingone such technique comfortably.

M3 - Conference contribution

SP - 155

EP - 164

BT - Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

PB - Association for Computational Linguistics

ER -

Padmanabhan D, Visweswariah K. Unsupervised Solution Post Identification from Discussion Forums. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Computational Linguistics. 2014. p. 155-164

Access to Document