Fashionista: A Clothing Application with Collaborative Filtering Based Recommendation System
Sahil Birwadkar, Rutuja Firke, Vidula Raje, Ruchira Vaity, Archana Vasant Mire · Journal of Emerging Technologies and Innovative Research · 2021
Recommender systems are used by various e-commerce sites to recommend different products to their customers. The products can be recommended based on the ratings given by the customers and various other implicit parameters, like purchase history or no of times a particular item was viewed. The recommendations in this project will be based on explicit feedback in the form of ratings. We use Matrix Factorization, which is a technique that discovers latent features and complex hidden relationship patterns among users or items with similar preferences or features, given a matrix to decompose. Gradient Descent is used to optimize the cost function. The e-commerce application will enable users to perform all the essential activities like rating, buying or adding a product to the wishlist and cart. Users can browse for items or explore the recommendations made to them.