DETERMINING MOVIE RANKINGS USING NON-PERSONALIZED AND PERSONALIZED APPROACHES FOR RECOMMENDER SYSTEM

• Main Author: Izzah, NUrul
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/54924
• Institution: Institut Teknologi Bandung

Providing a useful suggestion of products to online users to increase their consumption on websites is the goal of many companies nowadays. People usually select or purchase a new product based on some friend’s recommendations, comparison of similar products, or feedback from other users. In order to do all these tasks automatically, a recommender system must be implemented. The recommender systems are tools that provide suggestions that best suit the client’s needs, even when they are not aware of it. Recommender systems are beneficial to both service providers and users. They have proved to improve the decision-making process and quality. In an e-commerce setting, recommender systems enhance revenues, for the fact that they are effective means of selling more products. One of the industries that this final project wants to highlight is the streaming service industry. It is no secret that these services will change the movie industry someday, therefore these services have to provides a better system that would attract the customers more. In this project, a movie recommendation system will be built from a dataset that is obtained through the TMDB API which is in correspondence with the movies that are listed in the MovieLens. The latest dataset comprises 100,000 ratings and 1,300 tag applications applied to 9000 movies by 700 users. The main approaches of recommender algorithms are the non-personalized and personalized approaches. The personalized one consists of several techniques which are content-based filtering, collaborative filtering, and the hybrid technique. All of them will be introduced in this final project. We will select the algorithms that best fit the data and we will implement and compare them. This final project concludes with having the hybrid recommender system is the most suitable recommender system to use. But on the other hand, other algorithms that are more complex can be developed to generate more a more precise recommender system.