Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal JIPI (Jurnal Ilmiah Penelitian dan Pembelajaran Informatika)
Mariana, Tiwuk
Unknown Affiliation
Computer Science & IT Education

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

Found 1 Documents
Search

 1 
DEEP LEARNING SOLUTION FOR SPARSITY PROBLEM TO IMPROVE RECOMMENDATION QUALITY Lestari, Sri; Mariana, Tiwuk
JIPI (Jurnal Ilmiah Penelitian dan Pembelajaran Informatika) Vol 11, No 1 (2026)
Publisher : STKIP PGRI Tulungagung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29100/jipi.v11i1.7027

Abstract

Recommendation systems have become indispensable across various platforms due to their ability to enhance personalized services. However, these systems face a critical challenge known as sparsity. Sparsity occurs when there are numerous gaps in data, making user preferences unknown. This results in less relevant recommendations, reducing system effectiveness and diminishing user satisfaction. Moreover, it can lead to missed business opportunities. The purpose of this study is to address the sparsity problem using Deep Learning to enhance recommendation quality. The research stages include literature review (SLR), data collection from the Netflix Prize dataset obtained from kaggle.com, data preprocessing, Deep Learning implementation, testing, analysis, and conclusions. The stages of this study are conducted literature study (SLR), data collection, data preprocessing, Deep Learning implementation, testing and analysis, and conclusions. The method of this study is carried out data preprocessing and imputaion using several existing methods by using the Netflix Prize dataset, data taken from kaggle.com. The result of this study shows that the Deep Learning method is able to solve the sparsity problem to improve the quality of recommendations, because the experimental results states that the Root Mean Squared Error (RMSE) value is the smallest compared to the Matrix-Factorization, SVD, KNN and other methods.
Co-Authors Sri Lestari