The recommendation system is a system that can suggest information based on the results of observation of users’ desires to users. In this study, the recommendation system can be implemented into an online music player application by displaying song recommendations so that the application looks more personal to its users. The research method used to design this music recommendation system is a collaborative filtering by which the music recommendations for users are determined. The system produces a pretty good prediction when viewed from the MAE (Mean Absolute Error) score of 0.09639423292263861 and RMSE (Root Mean Squared Error) of 0.024737713540837314, meaning that the smaller the evaluation result is or close to 0, the more accurate it will be. The results of the MAE and RMSE calculations show that the prediction error rate is very small, so that they can be used as a parameter for determining music recommendations according to users’ needs.
Copyrights © 2022