Predictive Analytics for Sentiment Classification of Social Media Data Using Deep Neural Network

Abstract

A huge amount of user-generated data in the form of tweets or reviews on social media can be collected and analyzed for making informed decisions. This paper uses the novel deep learning model, namely the Elite Opposition-based Bat Algorithm for Deep Neural Network (EOBA-DNN) for performing polarity classification of the social media data. The proposed method includes three major steps, such as preprocessing, term weighting, and sentiment classification for identifying the polarity of the data. The results show that the EOBA-DNN outperforms other existing algorithms with improved accuracy for Sentiment Classification.