A Composite Technique for Creating Contemporary MRS using Association Rule Mining & CF

Abstract

The vast amount of information on the internet has made information available, but it has also made it difficult for users to choose the information that is necessary or interesting to them. To address this issue, recommender systems (RS) were developed to find relevant information using information filtering. Using RS, users may find the appropriate data from a vast collection. There are several types of RS, but those developed using collaborative filtering techniques have proven to be the most effective for a variety of issues. One of the most popular RS accessible is called the Movie Recommendation System (MRS). In this paper, suggestions will be made based on the shared features of user items. Both user objects and item objects are frequent in the movie recommendation system. In order to provide stronger suggestions, this paper integrates the collaborative filtering technique with association rule mining. By integrating collaborative filtering with association rule mining, a hybrid strategy that takes use of both techniques' advantages can boost the recommendation system's performance. Consequently, the recommendations that were generated can be regarded as strong recommendations. Collaborative filtering uses the past behavior of users to make recommendations, while association rule mining looks for patterns in the data to identify items that are frequently bought together. Combining these two approaches can help overcome the limitations of each individual method, such as the need for a large amount of data for collaborative filtering or the lack of personalization in association rule mining. This paper combines data mining and conventional filtering techniques to provide movie recommendation suggestions.