Sentiment Classification of Reviews Using Combination of Ant Colony Optimization and ANN

With the availability of many social media sites, the reviews and comments by the customers are freely available for analysis. As reviews are in text, it contains a large volume of words. The amount of words needs to be reduced without affecting the result of the analysis for future processing work. Thus, the Ant Colony Optimization (ACO) is used to select the reviews. For selection of the reviews, a fitness function is found out, that evaluate the fitness of each reviews, the reviews pass the fitness test will marked as selected reviews. These selected reviews are given input to Artificial Neural Network (ANN) for further processing work. The final output of the process is evaluated with confusion matrix and other parameters to validate the work with the previous paper work. The proposed approach has shown an accuracy of 97.5%, which is comparably better than existing paper.