Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal JURIKOM (Jurnal Riset Komputer)
Muhammad Daffa Ginting
Universitas Islam Negeri Sumatera Utara, Medan
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Analisis dan Implementasi Singular Value Decomposition (SVD) untuk Sistem Rekomendasi Anime Berbasis Collaborative Filtering Muhammad Daffa Ginting; Muhammad Ikhsan
JURIKOM (Jurnal Riset Komputer) Vol. 13 No. 1 (2026): Februari 2026
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/jurikom.v13i1.9462

Abstract

The rapid growth of digital platforms has significantly increased the number of available anime titles, making it difficult for users to choose content that matches their preferences. In addition, the diversity of genres and content characteristics accessible to different age groups highlights the need for a system that can provide recommendations more accurately and efficiently. This study aims to develop an anime recommendation system based on collaborative filtering by applying the truncated Singular Value Decomposition (SVD) method to address the data sparsity problem in the user–item matrix. The dataset was collected from the AniList platform and consists of 30 users, 100 anime titles, and 1,230 explicit rating records. The evaluation was conducted using a hold-out scheme with an 80% training set (992 ratings) and a 20% testing set (248 ratings). Prediction performance was measured using an RMSE of 1.6464 and an MAE of 1.2370, while the quality of Top-N recommendations was assessed using Precision@10. The experimental results indicate that the best configuration was achieved at k = 4, which produced the lowest prediction error on the test set and generated relevant recommendations. The average Precision@10 obtained was 0.1897, meaning that approximately 18.97% of the Top-10 recommendations provided by the system were considered relevant.
Co-Authors Muhammad Ikhsan