Please use this identifier to cite or link to this item: https://hdl.handle.net/11147/5119
Title: New features for sentiment analysis: Do sentences matter?
Authors: Gezici, Gizem
Yanıkoğlu, Berrin
Tapucu, Dilek
Saygın, Yücel
Tapucu, Dilek
Izmir Institute of Technology. Computer Engineering
Keywords: Machine learning
Polarity detection
Sentiment analysis
Sentiment classification
Issue Date: 2012
Publisher: CEUR Workshop Proceedings
Source: Gezici, G., Yanıkoğlu, B., Tapucu, D., and Saygın, Y. (2012, September). New features for sentiment analysis: Do sentences matter?. Paper presented at the Proceedings of the 1st International Workshop on Sentiment Discovery from Affective Data (SDAD 2012), Bristol, UK.
Abstract: In this work, we propose and evaluate new features to be used in a word polarity based approach to sentiment classification. In particular, we analyze sentences as the first step before estimating the overall review polarity. We consider different aspects of sentences, such as length, purity, irrealis content, subjectivity, and position within the opinionated text. This analysis is then used to find sentences that may convey better information about the overall review polarity. The TripAdvisor dataset is used to evaluate the effect of sentence level features on polarity classification. Our initial results indicate a small improvement in classification accuracy when using the newly proposed features. However, the benefit of these features is not limited to improving sentiment classification accuracy since sentence level features can be used for other important tasks such as review summarization.
Description: 1st International Workshop on Sentiment Discovery from Affective Data 2012, SDAD 2012 - In Conjunction with ECML-PKDD 2012; Bristol; United Kingdom; 28 September 2012 through 28 September 2012
URI: http://hdl.handle.net/11147/5119
ISSN: 1613-0073
1613-0073
Appears in Collections:Computer Engineering / Bilgisayar Mühendisliği
Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection

Files in This Item:
File Description SizeFormat 
5119.pdfConference Paper145.95 kBAdobe PDFThumbnail
View/Open
Show full item record

CORE Recommender

Page view(s)

4
checked on Jul 25, 2021

Download(s)

2
checked on Jul 25, 2021

Google ScholarTM

Check


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