Using Recommendation Engines to Improve the userexperience in a Cloud Analytic Solution

Abstract: Recommender systems have become an integral part of everyday human life because of tworeasons: addressing the issue of information filtering in the world of big data which limits therecommendation engine’s capabilities, and improving user experience by helping the userswhat users want. It is very often that users are unable to find preferred results with a simplequery or are unaware that even if it exits. The same case applied here to developrecommendation systems that provide insights recommendations to the users in order toimprove users’ experience. The purpose of the project was to develop and evaluate recommender systems with variousalgorithms to evaluate the best performing recommender system technique. The evaluation foreach of the recommender system was done using various metrics, and identifying thealgorithms that perform best in three techniques of recommender systems. It was observed thatthe Hybrid recommender system performed well with Matrix factorization, KNN and cosinesimilarity algorithms. The results show that the performance of hybrid recommender system matrix is best withfactorization and KNN. Therefore, it is suggested that the recommender system for the project isHybrid recommender system with matrix factorization. With this technique other algorithms canbe aligned to improve the performance of the recommender system.