Recommender System with Apache Spark

The rapid development of data has greatly impacted the storage, analysis, and processing performance of data on systems that use machine learning algorithms. Therefore, the aim of this research is to examine commonly used big data platforms and related machine learning libraries as well as classify recommendation systems. From there, we experimentally evaluate the accuracy of the recommendation system based on model-based collaborative filtering using Alternating Least Squares—ALS in Spark's Machine Learning Library (MLlib) for the 1 M MovieLens dataset. The experiment produces good results when evaluating the accuracy of the recommendation system based on Spark’s MLlib. Additionally, the report also examines changes in the precision of the recommender engine based on adjusting the parameters of the ALS algorithm.