Mining Experts in Technical Online Forums

Many organizations use or host discussion lists, in the form of online forums and email lists. Analyzing the content of those discussion lists is an effective solution to the task of expert finding, since experts tend to participate often by giving advice, and receive the best feedback. We present...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Das Neves, Fernando, Wasylyszyn, Fernando
Formato: Objeto de conferencia
Lenguaje:Inglés
Publicado: 2010
Materias:
Ciencias Informáticas
expert finding
discussions
polarity mining
machine learning
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/152668
http://39jaiio.sadio.org.ar/sites/default/files/39jaiio-asai-17.pdf
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-152668
record_format dspace
spelling I19-R120-10915-1526682023-05-09T20:05:01Z http://sedici.unlp.edu.ar/handle/10915/152668 http://39jaiio.sadio.org.ar/sites/default/files/39jaiio-asai-17.pdf issn:1850-2784 Mining Experts in Technical Online Forums Das Neves, Fernando Wasylyszyn, Fernando 2010 2010 2023-05-09T16:32:52Z en Ciencias Informáticas expert finding discussions polarity mining machine learning Many organizations use or host discussion lists, in the form of online forums and email lists. Analyzing the content of those discussion lists is an effective solution to the task of expert finding, since experts tend to participate often by giving advice, and receive the best feedback. We present a novel method to identify positive comments that helps to identify experts by combining author statistics with polarity mining. Our method is able to distinguish experts from flamers and other people that simply participates frequently in discussions. We demonstrate the validity of our approach by evaluating it with an online discussion forum in Spanish. Sociedad Argentina de Informática e Investigación Operativa Objeto de conferencia Objeto de conferencia http://creativecommons.org/licenses/by-nc-sa/4.0/ Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) application/pdf 187-198
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
expert finding
discussions
polarity mining
machine learning
spellingShingle Ciencias Informáticas
expert finding
discussions
polarity mining
machine learning
Das Neves, Fernando
Wasylyszyn, Fernando
Mining Experts in Technical Online Forums
topic_facet Ciencias Informáticas
expert finding
discussions
polarity mining
machine learning
description Many organizations use or host discussion lists, in the form of online forums and email lists. Analyzing the content of those discussion lists is an effective solution to the task of expert finding, since experts tend to participate often by giving advice, and receive the best feedback. We present a novel method to identify positive comments that helps to identify experts by combining author statistics with polarity mining. Our method is able to distinguish experts from flamers and other people that simply participates frequently in discussions. We demonstrate the validity of our approach by evaluating it with an online discussion forum in Spanish.
format Objeto de conferencia
Objeto de conferencia
author Das Neves, Fernando
Wasylyszyn, Fernando
author_facet Das Neves, Fernando
Wasylyszyn, Fernando
author_sort Das Neves, Fernando
title Mining Experts in Technical Online Forums
title_short Mining Experts in Technical Online Forums
title_full Mining Experts in Technical Online Forums
title_fullStr Mining Experts in Technical Online Forums
title_full_unstemmed Mining Experts in Technical Online Forums
title_sort mining experts in technical online forums
publishDate 2010
url http://sedici.unlp.edu.ar/handle/10915/152668
http://39jaiio.sadio.org.ar/sites/default/files/39jaiio-asai-17.pdf
work_keys_str_mv AT dasnevesfernando miningexpertsintechnicalonlineforums
AT wasylyszynfernando miningexpertsintechnicalonlineforums
_version_ 1765660137755770880

Ejemplares similares