Impact of Lexical Features on Answer Detection Model in Discussion Forums

Online forums have become the main source of knowledge over the Internet as data are constantly flooded into them. In most cases, a question in a web forum receives several responses, making it impossible for the question poster to obtain the most suitable answer. Thus, an important problem is how t...

Full description

Saved in:
Bibliographic Details
Main Authors: Atif Khan, Muhammad Adnan Gul, Abdullah Alharbi, M. Irfan Uddin, Shaukat Ali, Bader Alouffi
Format: Article
Language:English
Published: Wiley 2021-01-01
Series:Complexity
Online Access:http://dx.doi.org/10.1155/2021/2893257
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849683469163036672
author Atif Khan
Muhammad Adnan Gul
Abdullah Alharbi
M. Irfan Uddin
Shaukat Ali
Bader Alouffi
author_facet Atif Khan
Muhammad Adnan Gul
Abdullah Alharbi
M. Irfan Uddin
Shaukat Ali
Bader Alouffi
author_sort Atif Khan
collection DOAJ
description Online forums have become the main source of knowledge over the Internet as data are constantly flooded into them. In most cases, a question in a web forum receives several responses, making it impossible for the question poster to obtain the most suitable answer. Thus, an important problem is how to automatically extract the most appropriate and high-quality answers in a thread. Prior studies have used different combinations of both lexical and nonlexical features to retrieve the most relevant answers from discussion forums, and hence, there is no standard/general set of features that could be effectively used for relevant answer/reply post classification. However, this study proposed an answer detection model that is exclusively relying on lexical features and employs a random forest classifier for classification of answers in discussion boards. Experimental results showed that the proposed answer detection model outperformed the baseline technique and other state-of-the-art machine learning algorithms in terms of classification accuracy on benchmark forum datasets.
format Article
id doaj-art-d7cc9954849940fdb1e27c4b18d8b1c7
institution DOAJ
issn 1076-2787
1099-0526
language English
publishDate 2021-01-01
publisher Wiley
record_format Article
series Complexity
spelling doaj-art-d7cc9954849940fdb1e27c4b18d8b1c72025-08-20T03:23:51ZengWileyComplexity1076-27871099-05262021-01-01202110.1155/2021/28932572893257Impact of Lexical Features on Answer Detection Model in Discussion ForumsAtif Khan0Muhammad Adnan Gul1Abdullah Alharbi2M. Irfan Uddin3Shaukat Ali4Bader Alouffi5Department of Computer Science, Islamia College Peshawar, Peshawar, KP, PakistanDepartment of Computer Science, Islamia College Peshawar, Peshawar, KP, PakistanDepartment of Information Technology, College of Computers and Information Technology, Taif University, P.O. Box 11099, Taif 21944, Saudi ArabiaInstitute of Computing, Kohat University of Science and Technology, Kohat, PakistanDepartment of Computer Science, Islamia College Peshawar, Peshawar, KP, PakistanDepartment of Computer Science, College of Computers and Information Technology, Taif University, P.O. Box 11099, Taif 21944, Saudi ArabiaOnline forums have become the main source of knowledge over the Internet as data are constantly flooded into them. In most cases, a question in a web forum receives several responses, making it impossible for the question poster to obtain the most suitable answer. Thus, an important problem is how to automatically extract the most appropriate and high-quality answers in a thread. Prior studies have used different combinations of both lexical and nonlexical features to retrieve the most relevant answers from discussion forums, and hence, there is no standard/general set of features that could be effectively used for relevant answer/reply post classification. However, this study proposed an answer detection model that is exclusively relying on lexical features and employs a random forest classifier for classification of answers in discussion boards. Experimental results showed that the proposed answer detection model outperformed the baseline technique and other state-of-the-art machine learning algorithms in terms of classification accuracy on benchmark forum datasets.http://dx.doi.org/10.1155/2021/2893257
spellingShingle Atif Khan
Muhammad Adnan Gul
Abdullah Alharbi
M. Irfan Uddin
Shaukat Ali
Bader Alouffi
Impact of Lexical Features on Answer Detection Model in Discussion Forums
Complexity
title Impact of Lexical Features on Answer Detection Model in Discussion Forums
title_full Impact of Lexical Features on Answer Detection Model in Discussion Forums
title_fullStr Impact of Lexical Features on Answer Detection Model in Discussion Forums
title_full_unstemmed Impact of Lexical Features on Answer Detection Model in Discussion Forums
title_short Impact of Lexical Features on Answer Detection Model in Discussion Forums
title_sort impact of lexical features on answer detection model in discussion forums
url http://dx.doi.org/10.1155/2021/2893257
work_keys_str_mv AT atifkhan impactoflexicalfeaturesonanswerdetectionmodelindiscussionforums
AT muhammadadnangul impactoflexicalfeaturesonanswerdetectionmodelindiscussionforums
AT abdullahalharbi impactoflexicalfeaturesonanswerdetectionmodelindiscussionforums
AT mirfanuddin impactoflexicalfeaturesonanswerdetectionmodelindiscussionforums
AT shaukatali impactoflexicalfeaturesonanswerdetectionmodelindiscussionforums
AT baderalouffi impactoflexicalfeaturesonanswerdetectionmodelindiscussionforums