There are a diverse set of products for a particular type on the internet. When any user tries to find out best product among a certain type it is very much difficult to do it manually go through every one of them. That's why manually searching is not very efficient. In that scenario, recommendation system plays a great important role to recommend the best products. In this study, we develop a recommendation system for the organization that works with movies. Our recommendation system recommends movies based on user data. It takes the users data from the user's activity and based on that data it recommends the movies to the user. When our recommender system tries to recommend the movies to the user it heavily depends on the weight of the movies. These weighted values of movies aren't just random. It has a strong correlation with user's data or preference which we collected from user's activity. We correlate user's data and weight with a certain formula. This weighted value helps us to recommend movies to the user. Our recommendation system internally used k-means algorithm. Which we applied on to those weighted value to form clusters of movies and we recommend the cluster of movies to the user which has a highest mean movie rating.
展开▼