A Comparative Study of Machine Learning Algorithms using Feature Selection Methods for Movie Review Analysis

Volume 11 Issue 3 January - March 2017

Research Paper

A Comparative Study of Machine Learning Algorithms using Feature Selection Methods for Movie Review Analysis

Rajwinder Kaur*, Prince Verma**

* M.Tech Scholar, Department of Computer Science and Engineering, CT Institute of Engineering Management and Technology, Punjab, India.

** Assistant Professor, Department of Computer Science and Engineering, CT Institute of Engineering Management and Technology, Punjab, India.

Kaur, R., and Verma, P. (2017). A Comparative Study of Machine Learning Algorithms using Feature Selection Methods for Movie Review Analysis. i-manager’s Journal on Software Engineering, 11(3), 1-9. https://doi.org/10.26634/jse.11.3.13620

Abstract

Nowadays, the analysis of social sites, such as movie reviews’ sites, facebook, news feeds, and online shopping sites has been a broad area of research and customers post a large number of reviews in the form of comments to reveal their feelings as well as opinions as positive, negative, or neutral about a particular movie, product, pictures etc. To predict the reviews of users of such websites is a complex decision making process. These types of sites help the people to take decision about products. This paper proposes a Random Forest classifier with Information Gain based feature selection method for classification of movie review datasets. The results show that Information Gain method with Random Forest classifier has better performance in terms of Accuracy, Precision, and Recall.