Fake product reviews monitoring and removal using Machine learning

The rapid growth of e-commerce platforms has led to an increase in fake product reviews, which mislead consumers and manipulate product ratings. This paper presents an automated system for fake product review detection and removal using Machine Learning (ML). The system classifies reviews as genuine or fake based on textual features, user behaviour, and review credibility scores. The proposed approach utilizes Natural Language Processing (NLP) techniques such as TF-IDF, BERT, and Word2Vec for feature extraction, and classification models including Random Forest, Support Vector Machine (SVM), and Long Short-Term Memory (LSTM) networks for predictive analysis. The dataset is pre-processed using tokenization, stop-word removal, and stemming, ensuring high-quality input for the models. To enhance reliability, the system integrates community feedback, where low-rated reviews are flagged for potential removal. Additionally, an admin panel is provided for manual intervention in disputed cases.[3] The system is developed using Flask/Django for the backend, PostgreSQL for data storage, and a React-based dashboard for visualization and review management. Performance evaluation metrics such as accuracy, precision, recall, and F1-score are used to validate the effectiveness of the proposed method. Initial experimental results demonstrate a significant improvement in fake review detection accuracy compared to traditional rule-based approaches.