Movies recommendation based on user interaction with social network

Abstract

Recommendation systems (RS) play an important role in artificial intelligence and help make a system capable of choosing what is right for users. The personalization of information is the key that has made recommendation systems at the highest level, for this we have realized a system that uses the user's personal information to process the case of a new user and make a personalized recommendation. In this project a new idea for recommending movies has been created, this system is based on user interaction with tweeter, it is designed based on the similarity between users tweets and movies summaries, by finding this type of resemblance we can discover personal interests for each user and we can determine whether the user like a particular item or not only by computing the similarity. After collecting those information and decide which movies a new user like, the recommendation part comes a head by implementing matrix factorization approach for recommendation systems, basically when a new user who have no interaction with any item, the quality of recommendation will be at its worst. But by discovering the movies that user like based on the similarity calculation the system will be having information’s about user personal interests and then the quality of the recommendation could be better by using user interaction with tweeter.