Revealing the Reliability of Amazon Products via Innovative Fake Review Detection using Machine Learning

Purchases on e-commerce sites, like Amazon, are highly impacted by customer reviews. On the negative side, the rise of fake reviews is a risk for products and services. This study proposes an innovative approach for automatic review spam detection that integrates supervised and deep learning methodologies using LSTM networks and BERT. The approach combines NLP, sentiment analysis, and feature extraction based on meta-data to improve classification performance. The comparison with traditional machine learning classifiers, such as Logistic Regression, Random Forest, and Support Vector Machines (SVM), proves the efficacy of artificial intelligence techniques in detecting fraudulent reviews. The results highlight the effectiveness of artificial intelligence in solving review fraud and restoring confidence in online shopping.