Sentiment Analysis Approach Based on N-gram and KNN Classifier

Abstract

Sentiment analysis is the approach which is designed to analyze positive, negative and neutral aspects of any text unit. In the past years, many techniques were designed for the sentiment analysis of twitter data. Based on the previous study about sentiment analysis, a novel approach is presented in this research paper for the sentiment analysis of twitter data. The proposed approach is the combination of feature extraction and classification techniques. N-gram algorithm is applied for the feature extraction and KNN classifier is applied to classify input data into positive, negative and neutral classes. To validate the proposed system, performance is analyzed in terms of precision, recall and accuracy. The results of the experiment of proposed system show that it performs well as compared to the existing system which is based on SVM classifier.