1. UKM Learning and Research Repository
  2. Journal Article / Artikel Jurnal
  3. Journal Content Pages/ Kandungan Halaman Jurnal
Please use this identifier to cite or link to this item: https://ptsldigital.ukm.my/jspui/handle/123456789/578542
Full metadata record
DC FieldValueLanguage
dc.contributor.authorNurfadhlina Mohd Sharef (UPM)
dc.contributor.authorRozilah Rosli (UPM)
dc.date.accessioned2023-11-06T03:03:22Z-
dc.date.available2023-11-06T03:03:22Z-
dc.date.issued2017-06
dc.identifier.issn0128-7680
dc.identifier.otherukmvital:116008
dc.identifier.urihttps://ptsldigital.ukm.my/jspui/handle/123456789/578542-
dc.descriptionSentiment analysis classification has been typically performed by combining features that represent the dataset at hand. Existing works have employed various features individually such as the syntactical, lexical and machine learning, and some have hybridized to reach optimistic results. Since the debate on the best combination is still unresolved this paper addresses the empirical investigation of the combination of features for product review classification. Results indicate the Support Vector Machine classification model combined with any of the observed lexicon namely MPQA, BingLiu and General Inquirer and either the unigram or inte-gration of unigram and bigram features is the top performer.
dc.language.isoen
dc.publisherUniversiti Putra Malaysia Press
dc.relation.haspartPertanika Journals
dc.relation.urihttp://www.pertanika.upm.edu.my/regular_issues.php?jtype=2&journal=JST-25-S-6
dc.rightsUKM
dc.subjectProduct review
dc.subjectSentiment classification
dc.subjectSentiment features
dc.titleEmpirical investigation of feature sets effectiveness in product review sentiment classification
dc.typeJournal Article
dc.format.volume25
dc.format.pages125-132
dc.format.issueSpecial Issue
Appears in Collections:Journal Content Pages/ Kandungan Halaman Jurnal

Files in This Item:
File Description SizeFormat 
ukmvital_116008+Source01+Source010.PDF446.4 kBAdobe PDFThumbnail
View/Open
Show simple item record Recommend this item


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.