A Framework for Movie Recommendation Using Content-Based Filtering

In our today’s life items served to users on the basis on their likings play a vital role to boost up the sales and in this regard recommender systems play an important role. As with the ease of availability of internet we are flooded with data so it is the necessity to deal with the correct data in order to generate fruitful results. The present paper proposes a framework with content based filtering using TF-IDF. Although a number of work on recommender systems have been done still everyday new innovation in the field of recommender system is done. An important parameter for recommendation in our proposed approach is genre. Some of the similarity coefficient used by our model are Euclidean, Cosine, Pearson Correlation etc. Our proposed system is evaluated with kNN to generate accuracy of 93.25%. Using k-Nearest Neighbors (KNN) for movie recommendation in machine learning has its own set of advantages as KNN is a simple and easy-to-understand algorithm. Its basic idea involves finding similarities between data points based on their features. KNN can be used for both user-based and item-based collaborative filtering. KNN can be used in online recommendation systems where real-time recommendations are needed. The paper is organized into five prominent sections. Section 1 comprise of introduction, Sect. 2 discusses the related work, Sect. 3 discuss the methodology used, Sect. 4 comprise experimental result and finally Sect. 5 contains the conclusion of presented work.