A Study and Analysis of Issues and Attacks Related to Recommender System

Abstract

Information and communication technologies (ICTs) currently play a role in pervasive assistive tools in our daily life. Their usage is fast growing in various domains like healthcare, e-learning, e-government, e-tourism, banking sector, etc. Nowadays, we are living in the information age and entering the age of recommendation. Recommender systems are intelligent information retrieval systems which help in finding relevant information from a large information space based on the interest of the users. In ICT, the recommender system helps in finding accurate services/items from the large space of items. The recommender system is widely accepted and the most important approach that provides the solution for information overhead problem. Even though user spends a lot of time and effort for finding essential and expected information, it may result in lower precision and recall rates. Nowadays, users spend the shortest span of time and want to obtain the best possible items and services. At present, many recommender systems have been developed in several contexts, but these are not sufficient to accomplish the user needs. So, it is essential to develop the highly featured recommender system. For developing and designing such type of systems, many issues and challenges need appropriate attention. The main purpose of this chapter is to investigate and illustrate the current challenges, issue attacks, and applications in recommender systems. In this chapter, we also propose a recommender system model based on collaborative filtering and intention mining where intentions are generated using gender and age information of the users.