Systematic Review on Machine Learning based Recommendation System

Recommender systems are a technology that helps in choosing items that are relevant from a large number of items. Incorporating recommender systems across these applications allows brands to tailor their offerings, optimize user engagement, and ultimately increase customer satisfaction and loyalty by personalizing product recommendations based on user profiles or search recommendations based on past behavior. This paper is a systematic review of the various research trends that reflect the development and improvement of the recommender model over time. Every recommender system model is developed either using different mathematical models or machine learning algorithms. The involvement of Machine Learning Models has improved the performance of recommender systems but these recommender systems persist with problems and challenges remain in the form of data scarcity, cold-start, scalability, time consumption, and accuracy causing a decline in the quality of the systems. This paper contributes significantly by providing the current state of persisting recommender systems that are analyzed through several research papers.