Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal Media Journal of General Computer Science (MJGCS)
M Nabil Fadhlurrahman
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Optimizing Amazon Reviews Using Principal Component Analysis, Feature Selection On Random Forest Classifier M Nabil Fadhlurrahman; Mutiara Yudina Fitrah
Media Journal of General Computer Science Vol. 2 No. 1 (2025): MJGCS
Publisher : MASE - Media Applied and Engineering

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62205/mjgcs.v2i1.35

Abstract

Dataset optimization is an important step in machine learning to improve model performance. This review discusses the use of Random Forest, Principal Component Analysis (PCA), and Feature Selection algorithms to optimize datasets. Based on this review, the combination of Random Forest, PCA, and Feature Selection is proven to be effective in improving machine learning model performance. This combination can help reduce overfitting, improve prediction accuracy, and speed up the model training process. In our experiments with the Amazon Reviews dataset, this optimized approach achieved an impressive accuracy of 91%, demonstrating a significant improvement over baseline models.
Optimization of Heart Failure Risk Prediction Using Random Forest Classifier Algorithm M Nabil Fadhlurrahman; Winanto, Eko Arip
Media Journal of General Computer Science Vol. 2 No. 2 (2025): MJGCS
Publisher : MASE - Media Applied and Engineering

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62205/mjgcs.v2i2.105

Abstract

This study discusses the optimization of heart failure prediction using the Random Forest Classifier algorithm with a focus on feature selection marked by a threshold and the number of features used. The results of the analysis show that the right threshold has a significant effect on model performance. At a threshold of 0.02, the model achieves the best performance with the highest accuracy, precision, and F1-score values. However, increasing the threshold above 0.08 causes a gradual decrease in model performance. In addition, the number of features used also affects the prediction results, where the right combination of features can increase the effectiveness of the classification. Therefore, this study emphasizes the importance of optimizing thresholds and feature selection in building more accurate and efficient prediction models.
Co-Authors Mutiara Yudina Fitrah