Fake Product Fake Detection

Inspired by the growing issue of counterfeit products on platforms like Amazon and Flipkart, we developed a Fake Product Detection System that helps users identify whether a product is genuine or fake before purchasing. Our system analyzes customer reviews using NLP techniques, examines seller ratings, checks abnormal price deviations, and compares product images using deep learning models to generate a real/fake probability score. We built it by collecting and preprocessing product data, applying machine learning models such as Logistic Regression, Random Forest, and CNN for image validation, and deploying the solution through a simple web interface. During development, we faced challenges like limited labeled datasets, imbalanced data, and detecting realistic fake reviews, but we successfully created a functional prototype with strong classification performance. Through this project, we learned the importance of feature engineering, handling messy real-world data, integrating multiple AI techniques, and focusing equally on deployment and user experience.

Built With

• googlaistudio
• mern
• vscode