Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Journal of Information Systems and Informatics Tech : Journal of Engineering Science
Dhiyaussalam
Politeknik Negeri Banjarmasin
Automotive Engineering Chemical Engineering, Chemistry & Bioengineering Civil Engineering, Building, Construction & Architecture Computer Science & IT Electrical & Electronics Engineering Engineering

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

Found 2 Documents
Search

 1 
Impact of Feature Selection on the Performance of KNN and SVM in Heart Disease Prediction Dhiyaussalam; M. Helmy Noor; Herlinawati; Isna Wardiah
Tech : Journal of Engineering Science Vol 1 No 1 (2025): Pengembangan dan Penerapan Solusi Rekayasa untuk Tantangan Lingkungan, Industri,
Publisher : Yayasan Penelitian dan Pengabdian Masyarakat Sisi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.69836/tech.v1i1.353

Abstract

Feature selection plays a vital role in enhancing the performance of machine learning models by eliminating irrelevant or redundant attributes. This study investigates the impact of feature selection on the classification accuracy of K-Nearest Neighbors (KNN) and Support Vector Machine (SVM) in predicting heart disease. Using the UCI Heart Disease data, which initially includes 13 input features, feature importance scores were calculated using a Random Forest model. A threshold-based method was then applied to identify and retain the most informative features. Through iterative testing of importance thresholds, a value of 0.03 yielded the best results, reducing the feature set from 13 to 9 attributes. Classification models were trained and evaluated using full and reduced feature sets. Performance was assessed using accuracy, precision, recall, and F1-score and validated with 5-fold cross-validation. The results demonstrate significant performance gains after feature selection. The KNN classifier improved accuracy from 83% to 92%, with notable gains in recall and F1-score for the positive class. Similarly, SVM achieved 92% accuracy, with improved precision and overall performance stability. These findings suggest that data-driven feature reduction simplifies the model and enhances its predictive power. This study systematically compares feature selection effects on two distinct machine learning algorithms and offers practical insights for optimizing medical prediction models in clinical decision support systems.
A Lightweight Classical Machine Learning Pipeline for Rice NPK Deficiency Classification Using Hand-Crafted Feature Fusion Dhiyaussalam; Kun Nursyaiful Priyo Pamungkas; Wanvy Arifha Saputra; Ahmad Yusuf
Journal of Information System and Informatics Vol 8 No 2 (2026): April
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v8i2.1486

Abstract

Many high-accuracy deep learning solutions for plant nutrient deficiency remain impractical in resource-limited settings due to computational cost and limited explainability. This study proposes a lightweight classical machine learning pipeline for rice leaf NPK (nitrogen, phosphorus, potassium) deficiency classification on the publicly available Kaggle Nutrient-Deficiency-Symptoms-in-Rice dataset (1,156 images); all results should be interpreted in this dataset context rather than as field-validated performance. The pipeline applies HSV-based leaf segmentation to reduce background influence. It extracts a 126-dimensional feature set combining masked color moments, HSV histograms, vegetation indices, LBP and GLCM texture descriptors, and spatial symptom ratios. Hyperparameters are tuned via RandomizedSearchCV with 5-fold StratifiedKFold and macro-F1 scoring; final evaluation uses a held-out 80/20 stratified test set kept separate throughout tuning. XGBoost achieves the best test performance (accuracy 0.9267; macro-F1 0.9233), followed by SVM-RBF (0.9224; 0.9187) and Random Forest. Feature importance analysis confirms that color moments dominate class separability, with texture and spatial features providing complementary support. The dominant remaining error is phosphorus–potassium confusion. The novelty lies in integrating leaf-focused preprocessing with a structured, low-cost feature representation suitable for mobile or edge deployment.
Co-Authors Ahmad Yusuf Herlinawati Kun Nursyaiful Priyo Pamungkas M. Helmy Noor Wanvy Arifha Saputra Wardiah, Isna