Classification Technique for Sentiment Analysis of Twitter Data

Abstract

The sentiment analysis is the technique which can analyze the behavior of the user. The data which is analyzed is the twitter data. The four steps are followed for the sentiment analysis in the first step, the first step is applied in which data pre-processed. In the second step feature of the data will be extracted which is given as input to the third step in which data is classified for the sentiment analysis. In this paper, pattern based technique is applied for the feature extraction in which patterns are generated from the existing patterns which increase the accuracy of data classification. The proposed algorithm is been implemented in python using the nltk tool box and it is been analyzed that execution time is reduced and accuracy is increased at steady rate.