Distributed processing using cosine similarity for mapping Big Data in Hadoop

The analysis of big data is an issue that has become very important in recent years. The use of algorithms for processing big data that have generated as a result valuable information to organizations can be considered one of the biggest developments and most important lines of work today. This paper aims to show results in implementing cosine similarity for mapping big data in a flat database. For this purpose the information from movie ratings will be used, so it will result in a recommendation of a movie highly related to another. If the information used for testing is considered real, these results could be useful for the development of a recommendation system for products and services from an organization which has as well the records of their customers' ratings.