Machine & Deep Learning Techniques for Detection of Fake Reviews: A Survey
Jane Crystal Rodrigues, Jane Trinity Rodrigues, Vialli Luis Keji Gonsalves, Aditya Uday Naik, Pratiksha Shetgaonkar, Shailendra Aswale
Abstract
Presently, review sites are frequently confronted with the spread of wrong information, this could be done by an individual spammer or group spammers who compose fake reviews to either advertise or demean certain products that are available. This paper focuses on the detection of these fake reviews using sentiment analysis. Various data pre-processing techniques are used to convert the reviews to the proper format for analysis and for detection. The proposed methodology is to analyze reviews by making use of neural networks such as LSTM, Bi-LSTM and GRNN, and the activation functions in them namely ReLU, Sigmoid, TanH and comparing them to find the optimal model for analysis.