Enhanced Fake Review Detection and Sentiment Analysis with Hybrid Deep Learning Models

In an era where online reviews significantly influence consumer decisions, ensuring the authenticity of the review is of paramount importance. This proposed work introduces an innovative solution through a hybrid deep learning model for detecting fake reviews and analyzing sentiment. For fake review detection, a Bidirectional Long Short-Term Memory (BI-LSTM) network captures complex textual dependencies, effectively identifying subtle patterns indicative of fraudulent reviews. Simultaneously, sentiment analysis is performed using a combination of Hybrid Convolutional Neural Network (H-CNN) and Bidirectional Gated Recurrent Unit (BI-GRU) models. The BI-GRU enhances contextual understanding for nuanced sentiment analysis, while the H-CNN excels in feature extraction, supporting sentiment classification. This hybrid architecture demonstrates robustness in distinguishing fake reviews and extracting sentiment nuances from text data, thus providing consumers with more reliable information for informed decision-making in the digital marketplace.