Collaborative Filtering on Movie Recommendation Using Big Data

Abstract

These days, digitalization is increasing with rapid growth of personal and by home digital devices and the daily usage of internet, we generate a very large amount of amount of data, termed as "Big data”. Movie recommendation systems can be enhanced to the needs of the users as individually. Collaborative filtering is a popular approach in big data domain to create recommendation systems. We describe a technique collaborative recommendation technique based on an algorithm specifically designed to mine association rules for this purpose. We use Alternating Least Squares. We use the Association rule mining approach to generate the rules to recommend movies to a user. We employ associations between users and association between the items. We build collaborative filtering in Apache spark. Apache Spark is the leading open- source unified analytics engine for big data processing. We use Euclidean distance similarity.