Sentiment Analysis on Movie Reviews Based on Combined Approach

Anurag Mulkalwar, Kavita Kelkar

Abstract: Sentiment Analysis is one of the most important area of concern towards classifying sentiments from any given textual information. Sentiment analysis can be performed on textual data by using various machine learning methods like supervised learning or unsupervised learning. But no any individual method is sufficient to classify sentiments with that much precision. In this paper we are going to propose new approach to perform sentiment analysis on movie review called as Combined Approach. It uses two separate classifier Support Vector Machine (SVM) and Hidden Markov Model (HMM). Then it combines results of these classifier using classifier combine rule.

Keywords: Sentiment Analysis, Combined Approach, HMM, SVM