Accuracy‐Assured Privacy‐Preserving Recommender System Using Hybrid‐Based Deep Learning Method

Recommender System is an efficient information filtering system which has been used in different fields to customize applications by predicting and recommending various items. Collaborative Filtering (CF) is most well‐known technique of recommender system which is used to find a new one among various items that correspond to user's choice by measuring similar users’ interest shown on other similar items. Recommender system is decision making system which is used in various fields to personalize applications by recommending different kinds of items. CF is a famous filtering technique in recommender system which is used in cross domain applications to predict and recommend an item to a particular user. Here Privacy and accuracy are two main factors which play major role for recommender system. There are different machine learning and deep learning based collaborative filtering methods used in recommender system. In this chapter, we propose Restrictive Boltzmann Machine Approach (RBM) and hybrid deep learning method i.e. RBM with Convolution neural network (CNN) (CRBM). These two proposed approaches provide better accuracy of the movie recommender system as compared to other existing methods. The proposed CRBM (RBM with CNN method) is best method which provides less mean absolute error (MAE) than all methods.