Sentiment Analysis for Twitter Stream Data by Combining Lexicon and Machine Learning Approaches

Abstract

Nowadays, Twitter Sentiment analysis has become popular as it helps the organization to determine marketing strategy by providing public opinions. Efficient techniques to collect a large amount of Twitter stream data and extracting sentiment information from collected raw data are essential demand. Traditional sentiment classification techniques do not perform well in Social Data. Acquiring effective training data is a challenge although learning based approaches are good for Social Data Sentiment Classification. Manual Labeling for training data is time and labor consuming. In this paper, Sentiment Analysis System for Twitter data is proposed with five modules: Data Collection, Preprocessing, Class Labeling, Classification Model Development and Sentiment Classification. The Sentiment Classification is implemented by combining lexicon and Supervised learning-based approaches. In this system, lexicon-based classifier is applied to label the class and suitable learning-based classifier is chosen for classification. Emoticon and slang words are considered for classification. To select suitable classifier, three different classification algorithms are evaluated. The performance evaluation shows that Naïve Bayes classifier is better and the proposed system can achieved the promising accuracy.