Optimization of K-Nearest Neighbor for Recommendation System

Abstract

A tool for interacting with vast and complex information spaces is a recommender system. They offer a tailored view of these areas, giving users the items, they are most likely to find interesting priority. The field has advanced significantly since 1995 in terms of the range of issues it addresses, the methods used, and the applications it can be put to use. Research on recommender systems incorporates a variety of artificial intelligence methods, such as constraint satisfaction, case-based reasoning, user modeling, machine learning, and data mining. Many online e-commerce platforms, including Amazon.com, Netflix, Pandora, and others, heavily rely on personalized recommendations. This wealth of realworld application knowledge has propelled researchers to broaden the application of recommendation systems to new and difficult domains. Because there is such a high demand for both online shopping and movies, businesses rely on it. Machine learning-based technology makes it easier than ever to locate our true target audience. Jobs that aid in our understanding of our requirements and offer suggestions for them are recommended. Items that consumers are looking for. In this study, various machine learning algorithms for recommending various product purchases are compared.