Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Dinamik
Gilang Febrianto
Universitas Teknokrat Indonesia
Computer Science & IT

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

Found 1 Documents
Search

 1 
Klasifikasi Tipe Anemia berdasarkan Parameter CBC menggunakan Metode SVM Gilang Febrianto; Erliyan Redy Susanto
Dinamik Vol 31 No 2 (2026)
Publisher : Universitas Stikubank

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35315/dinamik.v31i2.10586

Abstract

The precision medicine paradigm is shifting conventional diagnosis toward AI-based systems. However, the literature on Support Vector Machines (SVM) with systematic hyperparameter optimization for the multi-class classification of anemia (Iron Deficiency/ID, Iron Deficiency Anemia/IDA, Normal) remains limited. This study develops an optimized SVM model for identifying three categories of anemia using a Kaggle dataset (1,000 samples, 6 features: Hemoglobin, RDW, MCV, Age, Gender, Anemia Type). The methods include data preprocessing, an 80:20 stratified split, and hyperparameter optimization via 5-fold cross-validation (CV) grid search with exploration of kernels (linear, RBF, poly) and the C parameter (0.1, 1, 10, 100). Results show that an SVM with a linear kernel and C=100 achieves 100% accuracy on the test data, with perfect precision, recall, F1-score, and ROC AUC (1.000) for all classes, as well as a cross-validation mean accuracy of 99% (±1%). Feature analysis identifies Hemoglobin as the dominant predictor, followed by RDW, MCV, Age, and Gender. The study’s contributions include an SVM benchmark framework for anemia classification, a demonstration of the effectiveness of hyperparameter optimization, and the strengthening of the Indonesian medical informatics literature in the application of machine learning to metabolic diseases, which accelerates the digital transformation of clinical hematology practice.
Co-Authors Erliyan Redy Susanto