Item-Based Collaborative Filtering in Movie Recommendation in Real time

Abstract

Collaborative filtering is one of the most effective and adequate technique used in recommendation. The fundamental aim of the recommendation is to provide prediction of the different items in which a user would be interested in based on their preferences. Recommendation systems based on collaborative filtering techniques are able to provide approximately accurate prediction when there is enough data. User based collaborative filtering taechniques have been very powerful and success in the past to recommend the items based on user’s preferences. But, there are also some certain challenges such as scalability and sparsity of data which increases as the number of users and items increases. In a large website, it is difficult to find the interested information in a certain time. But the recommendation system filter out information and items that are best suitable for us. Although there are different recommendation approaches, yet collaborative filtering technique is very popular because of the effectiveness. In this work, movie recommender system has been described, which basically uses item-based technique of collaborative filtering to provide the recommendations of items, which is dynamic and will learn from the positive feedback.