Fake Product Identification Using Machine Learning

Abstract

Recognizing counterfeit goods can be difficult in some situations. If a person does not thoroughly inspect the product's details, it becomes simpler to create and sell counterfeit goods. For less tech-savvy clients who can scan the product with the use of a smartphone application to check the authenticity of the product received, this paper offers a superior alternative employing machine learning. The detection of logos (which includes both visual and textual representations) is the main focus. The model also includes the sentiment analysis of the product’s reviews. This technique is useful for predicting the validity of a product. The paper describes the Fake Product Identification Model developed using Convolution Neural Network (CNN) and Optical Character Recognition (OCR). This model determines whether a product is real or fake, and the user can make a wise decision before buying the product.