Grocery Product Classification and Recommendation System Based on Machine Learning and Customer Profile Identity

Abstract

In today’s world number of ecommerce companies is increasing day by day. Most of the ecommerce system allow the user to rate the product and also provide flexibility to view or submit reviews. Grocery Products are also sold online by few ecommerce giants, recommendations are provided to the end user based on collaborative filtering, content based or based on text review sentiments but there is no consideration of user likeness with respect to a particular food. In this paper the product classification is performed using both sentiment score computation based on combination of support vector machine and artificial neural network along with frequency computation on specific nutrition features namely salt, sugar, protein, energy and fat Once the customer registers into application a nutrition questioner is asked for customer and data analysis is performed based on user answers in order to classify the end user into a particular likeness category. There is a relationship established between the user and the products, the products are recommended based on high positive score, low negative score and high frequency under the user likeness category.