Prototype of a Recommendation System Based on Multi-agents in the Analysis of Movies Dataset

In this paper is made a proposal of a recommendation system based on multi-agents, showing the architecture designed, the server used for the development of the multi-agent system, as well as the communication between necessary agents to carry out a tour recommended. The implemented proposal allows to make suggestions to users about movies. By means of neural networks it is determined if the proposed route for the user is correct or if it is necessary to improve the suggestion for following recommendations. In order to generate the recommendations, the free dataset of MovieLens was used, where a database was created to allow the analysis of them; and also to obtain a response to new recommendations for users.