An Aspect based Sentimental Analysis Approach using Deep Neural Networks and Artificial Fish Swarm Optimization to Analyze the Sentiments in Tweets

Abstract

Sentimental Analysis is a technique that helps to gain insights from the 80% of unstructured data generated worldwide. Sentimental Analysis only detects sentiments present in the text while the Aspect Based Sentimental Analysis identifies the underlying features present in the sentiments. In the recent past, Deep learning algorithms have been widely applied for Sentimental Analysis. This paper endeavors to develop a DNN powered AFS algorithm based Sentiment Analysis model to identify the features present in the tweets. The AFS algorithm improves the functionality of aspect selection and eliminates the worst aspects present in the dataset. The AFS powered DNN can compute the relationship between the aspect and the polarity. The joint approach gives improved performance and provides optimization on the four datasets used in this paper. The experimental results demonstrate that the proposed model outperforms other classifiers by giving an overall accuracy of 99.31%.