Fake Review Detection Using Sentiment Analysis and Deep Learning

When it comes to determining whether or not to buy a product, reviews are very crucial. These review sites are continually bombarded with inaccurate information, which may be perpetrated by individuals or groups who fabricate fake reviews in order to promote or demonize specific goods. This research work concentrates on the identification of fake reviews or feedbacks using sentiment analysis and natural language processing. To transform the data into a format appropriate for analysis and detection, a pre-processing technique is used. The proposed method involves analyzing review feedbacks with deep learning neural networks like the Gated Recurrent Unit (GRU), Bidirectional LSTM (Bi-LSTM) & Long Short Term Memory (LSTM), and evaluating the results using activation functions like ReLu, TanH, and Sigmoid.