Publication Type : Journal Article
Publisher : International Journal of Applied Engineering Research, Research India Publications
Source : International Journal of Applied Engineering Research, Research India Publications, Volume 10, Number 17 (2015)
Url : https://www.scopus.com/inward/record.uri?eid=2-s2.0-84979584004&partnerID=40&md5=0cb18f4937a9e35661eb9a5944ab15ac
Campus : Coimbatore
School : School of Engineering
Center : Computational Engineering and Networking
Department : Electronics and Communication
Year : 2015
Abstract : To decide on anything in our day to day life, it is important to have an opinion. Every opinion has a sentiment which helps in carrying decisions easier. There is a huge amount of data on the web which needs to be mined in order to find its sentiment. This paper aims at classifying labelled textual Hindi movie reviews with different classifiers. The dataset has been segregated into positive and negative reviews before processing. The goal of this paper is to predict the sentiment of the online movie review which is in form of documents with varied size. A 10-fold-cross-validation is done in order to check the calibre of the classifier used. The test accuracy is checked using the F1 score considering both precision and recall. A detailed comparison of the unigram and bigram feature‟s accuracy of all the mentioned models is done. The proposed model is classified on the following classifiers Naïve Bayes, Logistic Regression and Random Kitchen Sink algorithm. Each one of these algorithms gave better accuracy when bigram was performed. Out of these four classifying algorithms, it is observed that Naive Bayes Multinomial model has the best accuracy with a 70.37%. Hence, this sentiment analysis model which is a developing big data application is suggested for industrial applications wherein predicting the sentiment is a vital component.
Cite this Research Publication : S. N. Vinithra, Dr. M. Anand Kumar, and Dr. Soman K. P., “Analysis of sentiment classification for Hindi movie reviews: A comparison of different classifiers”, International Journal of Applied Engineering Research, vol. 10, 2015.