Open Access Open Access  Restricted Access Subscription or Fee Access

Prediction of Fake Reviews and Review Sentiments using Sentiment Analysis

Sushma Dongre, M. Pal



As almost all products and services are available online hence e-commerce is growing rapidly. These online services provide the facility of writing the review about the product by the customers. Whenever any customer wants to buy the product, he believes in reviews given by other customers. But some spam reviews are given by individual of organization itself to increase the sale of product or competitor of other brand to degrade the sale. These spam reviews in turn mislead decision of customers. Sometime users blame the e-commerce site instead of manufacturer of product. Thus, we are proposing the system which does not allow fake reviews and we apply sentiment analysis (SA) to bifurcate the genuine reviews either as positive or negative. So, that manufacturer can improve the quality of product. We train the system using Naïve Bays algorithm to perform SA. SA is the technique under natural language processing (NLP).


Keywords: Fake review, Naïve Bays Algorithm, training, testing, NLP, SA

Full Text:




  • There are currently no refbacks.