Convolutional Neural Networks- Bidirectional Long Short Term Memory Based Fake Review Classification

Abstract

When it comes to deciding whether or not to purchase a product, online reviews have recently surpassed all other sources of information. Fake product reviews have become a lucrative business in recent years because they can either promote or degrade a product depending on how they are written. For many business owners, detecting fake reviews has been a hot topic of discussion. This study uses data preprocessing, feature extraction, and classification to detect fabricated product reviews, and the results are published. Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) are two common deep learning models that can discover complicated patterns in textual data. Long Short-Term Memory (LSTM) is a recurrent neural network with a tree topology that analyses variable-length sequential data. The bi-directional LSTM allows you to examine certain sequences from front to back as well as back to front. After extracting preprocessing information, a convolutional neural network with Bidirectional Long Short-Term Memory (CNN-BiLSTM) technology is utilized to identify and authenticate the review. At the end of the process, the effectiveness of the proposed strategy is assessed. The data demonstrate that the proposed strategy is effective regardless of the used goods or emotions.