leero-ady/big_data_2018

Product Recommendation System using Deep Learning

Recommender systems are an invaluable part of any online retailer. Helping the customer find all the items he wants as quickly as possible is essential for driving sales and ensuring positive customer feedback. Our project aims to utilise a wide range of machine learning algorithms to create a recommender system that predicts the products that the user wants to buy based on their shopping habits. We are implementing this using RNN - Recurrent Neural Network model with LSTM (Long Short-Term Memory).The fact that RNN with LSTM units can overcome problems of vanishing (or exploding) gradients is one of the reasons why we have chosen this model. We are implementing RNN with built-in tensorflow, numpy and few libraries. In the second level, we use Feedforward Nueral Network , which takes the internal representations(results) from the previous level as input. Keywords— Market Basket Analysis, Deep Learning, Neural Networks, Recommender System.