Show simple item record

dc.contributor.authorTrivedi, Shrawan Kumar
dc.contributor.authorDey, Shubhamoy
dc.contributor.authorKumar, Anil
dc.date.accessioned2019-01-17T10:22:47Z
dc.date.available2019-01-17T10:22:47Z
dc.date.issued2018-08-06
dc.identifier.citationShrawan Kumar Trivedi, Shubhamoy Dey, Anil Kumar, (2018) ‘Capturing user sentiments for online Indian movie reviews: A comparative analysis of different machine-learning models’, The Electronic Library, 36 (4), pp.677-695, DOI: 10.1108/EL-04-2017-0075en
dc.identifier.issn0264-0473
dc.identifier.doi10.1108/EL-04-2017-0075
dc.identifier.urihttp://hdl.handle.net/10545/623307
dc.description.abstractSentiment analysis and opinion mining are emerging areas of research for analysing Web data and capturing users’ sentiments. This research aims to present sentiment analysis of an Indian movie review corpus using natural language processing and various machine learning classifiers. In this paper, a comparative study between three machine learning classifiers (Bayesian, naïve Bayesian and support vector machine [SVM]) was performed. All the classifiers were trained on the words/features of the corpus extracted, using five different feature selection algorithms (Chi-square, info-gain, gain ratio, one-R and relief-F [RF] attributes), and a comparative study was performed between them. The classifiers and feature selection approaches were evaluated using different metrics (F-value, false-positive [FP] rate and training time).The results of this study show that, for the maximum number of features, the RF feature selection approach was found to be the best, with better F-values, a low FP rate and less time needed to train the classifiers, whereas for the least number of features, one-R was better than RF. When the evaluation was performed for machine learning classifiers, SVM was found to be superior, although the Bayesian classifier was comparable with SVM. This is a novel research where Indian review data were collected and then a classification model for sentiment polarity (positive/negative) was constructed.
dc.description.sponsorshipNAen
dc.language.isoenen
dc.publisherEmerald Insighten
dc.relation.urlhttps://www.emeraldinsight.com/doi/10.1108/EL-04-2017-0075en
dc.rightsArchived with thanks to The Electronic Libraryen
dc.subjectOpinion miningen
dc.subjectIndian movie reviewsen
dc.subjectMachine learning classifiersen
dc.subjectUser sentiment analysisen
dc.titleCapturing user sentiments for online Indian movie reviews.en
dc.typeArticleen
dc.contributor.departmentIndian Institute of Management Sirmaur, Sirmaur, Indiaen
dc.contributor.departmentIndian Institute of Management Indore, Indore, Indiaen
dc.contributor.departmentUniversity of Derbyen
dc.identifier.journalThe Electronic Libraryen
dc.contributor.institutionDepartment of IT and Systems, Indian Institute of Management Sirmaur, Sirmaur, India
dc.contributor.institutionDepartment of Information Systems, Indian Institute of Management Indore, Indore, India
dc.contributor.institutionDepartment of Decision Science, BML Munjal University, Gurgaon, India
dc.internal.reviewer-note15.01.2019 No attachment. Emailed author to provide. HLen
dc.dateAccepted2017-11-22
dc.dateAccepted2017-11-22
refterms.dateFOA2019-02-28T18:04:53Z
html.description.abstractSentiment analysis and opinion mining are emerging areas of research for analysing Web data and capturing users’ sentiments. This research aims to present sentiment analysis of an Indian movie review corpus using natural language processing and various machine learning classifiers. In this paper, a comparative study between three machine learning classifiers (Bayesian, naïve Bayesian and support vector machine [SVM]) was performed. All the classifiers were trained on the words/features of the corpus extracted, using five different feature selection algorithms (Chi-square, info-gain, gain ratio, one-R and relief-F [RF] attributes), and a comparative study was performed between them. The classifiers and feature selection approaches were evaluated using different metrics (F-value, false-positive [FP] rate and training time).The results of this study show that, for the maximum number of features, the RF feature selection approach was found to be the best, with better F-values, a low FP rate and less time needed to train the classifiers, whereas for the least number of features, one-R was better than RF. When the evaluation was performed for machine learning classifiers, SVM was found to be superior, although the Bayesian classifier was comparable with SVM. This is a novel research where Indian review data were collected and then a classification model for sentiment polarity (positive/negative) was constructed.


Files in this item

Thumbnail
Name:
_system_appendPDF_proof_hi(9).pdf
Size:
422.2Kb
Format:
PDF
Description:
Post print.

This item appears in the following Collection(s)

Show simple item record