Research Article
Movie Recommender System Comparison of User-based and Item-based Collaborative Filtering Systems
@INPROCEEDINGS{10.4108/eai.19-10-2018.2282541, author={Imam Dwicahya and P. H. Prima Rosa and Robertus Adi Nugroho}, title={Movie Recommender System Comparison of User-based and Item-based Collaborative Filtering Systems}, proceedings={Proceedings of the 1st International Conference on Science and Technology for an Internet of Things, 20 October 2018, Yogyakarta, Indonesia}, publisher={EAI}, proceedings_a={ICSTI}, year={2019}, month={4}, keywords={recommender system user-based collaborative filtering item-based collaborative filtering}, doi={10.4108/eai.19-10-2018.2282541} }- Imam Dwicahya
P. H. Prima Rosa
Robertus Adi Nugroho
Year: 2019
Movie Recommender System Comparison of User-based and Item-based Collaborative Filtering Systems
ICSTI
EAI
DOI: 10.4108/eai.19-10-2018.2282541
Abstract
Collaborative Filtering (CF) is a method that is widely used in recommendation systems. There are two approaches that are often used in CF, namely User-based CF and Item-based CF. The User-based CF approach requires several similar users to predict the rating of a new item. Meanwhile, Item-based CF requires several similar items to predict the rating of a new item. The number of similar users or similar items involved in predicting ratings will affect the computing load. This research aims to see the effect of the number of neighbors (similar user or similar items) used on the level of accuracy of rating predictions for an item. By using different numbers of neighbors for both User-based CF and Item-based CF, the results of the experiment show that the number of neighbors affects the level of accuracy although not too significant.