OPTIMIZATION OF MACHINE LEARNING ALGORITHMS IN THE CLASSIFICATION OF VECTOR-BORNE DISEASES - Research Report

Aug 4, 2026

Upload by Aurelia

Developing a predictive model is the study’s objective, targeting vector-borne diseases using multiple machine learning methods, including Random Forest, Logistic Regression, k-nearest Neighbors, Decision Tree, and XGBoost. Class imbalance is corrected through oversampling techniques such as SMOTE and Random Oversampling. The dataset is sourced from Kaggle and literature references from Frontiers in Ecology and Evolution, totaling about 9,490 entries with environmental, demographic, and clinical attributes. Dengue fever and Aedes aegypti are used as the case focus, with DT and XGBoost achieving the highest accuracy (99.2%).

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages7

Reading Time18 min

AccessFree to download

Introduction

Background on vector-borne diseases and surveillance challenges

Motivation: machine learning and class imbalance

Research objective and proposed approach

OPTIMIZATION OF MACHINE LEARNING ALGORITHMS IN THE CLASSIFICATION OF VECTOR-BORNE DISEASES Sukrul Ma’mun1; Eni Heni Hermaliani2* Magister Ilmu Komputer1, Sistem Informasi2 Universitas Nusa Mandiri, Jakarta, Indonesia 1,2 [https://nusamandiri.ac.id](https://nusamandiri.ac.id1)[1](https://nusamandiri.ac.id1),2 [14220019@nusamandiri.ac.id](14220019@nusamandiri.ac.id1)[1](14220019@nusamandiri.ac.id1), [enie_h@nusamandiri.ac.id](enie_h@nusamandiri.ac.id2)[2](enie_h@nusamandiri.ac.id2)*(*) Corresponding Author The creation is distributed under the Creative Commons Attribution-NonCommercial 4.0 International License. Abstract— Developing a predictive model is the objective of this study, focusing on vector-borne diseases using various machine learning methods, including Random Forest (RF), Logistic Regression (LR), k-nearest Neighbors (kNN), Tree (DT), and XGBoost. The main goal is to use oversampling techniques like SMOTE and Random Oversampling to correct the dataset's class imbalance. The dataset was obtained from Kaggle and literature references published in Frontiers in Ecology and Evolution (Endo and Amarasekare 2022), consisting of approximately 9,490 entries with environmental, demographic, and clinical attributes. Dengue Fever isone of the diseases that this study focuses on. Aedes aegypti mosquitoes spread it, and it is a significant health risk in tropical areas. The DT and XGBoost models had the highest accuracy, at 99.2%. Logistic Regression and Random Forest also did well, with 99.1% accuracy. KNN did well, too, but with a lower recall, at 99.0%. The ROC curve gave a complete picture of how well each model classified things. These findings indicate that when combined with proper data handling, machine learning models can significantly improve early detection of vector-borne diseases and support more accurate and timely decision-making in public health interventions. Keywords: Disease Control, Disease Prediction, Machine Learning (ML), SMOTE, Vector-Borne Diseases. Intisari—Mengembangkan model prediktif adalah tujuan dari penelitian ini, dengan fokus padapenyakit yang ditularkan melalui vektor menggunakan berbagai metode pembelajaranmesin, termasuk Random Forest (RF), Logistic Regression (LR), k-nearest Neighbors (kNN), Tree (DT), dan XGBoost. Tujuan utamanya adalah menggunakan teknik oversampling seperti SMOTE dan Random Oversampling untuk mengoreksi ketidakseimbangan kelas dataset. Dataset diperoleh dari Kaggle dan referensi literatur yang diterbitkandalam di Frontiers in Ecology and Evolution (Endo and Amarasekare 2022), yang terdiri dari sekitar 9.490 entri dengan atribut lingkungan, demografi, dan klinis. Demam Berdarah adalah salah satu penyakit yang menjadi fokus penelitian ini. Nyamuk Aedes aegypti menyebarkannya, dan merupakan risiko kesehatan yang signifikan di daerah tropis. Model DT dan XGBoost memiliki akurasi tertinggi, yaitu 99,2% . Logistic Regression dan Random Forest juga bekerja dengan baik, dengan akurasi 99,1% . KNN juga berhasil, tetapi dengan perolehankembali yang lebih rendah, yaitu 99,0% . Kurva ROC memberikan gambaran lengkap tentang seberapa baik setiap model mengklasifikasikan berbagai hal. Temuan ini menunjukkan bahwa biladikombinasikan dengan penanganan data yang tepat, model pembelajaran mesin dapat secara signifikan meningkatkan deteksi dini penyakit yang ditularkan melalui vektor dan mendukung pengambilan keputusanyang lebih akurat dantepat waktu dalam intervensi kesehatan masyarakat. Kata Kunci: Pengendalian Penyakit, Prediksi Penyakit, Pembelajaran Mesin (ML), SMOTE, Penyakit Tular Vektor. INTRODUCTION Vector-borne diseases, such as malaria and dengue fever, remain major public health threats, particularly in tropical regions like Indonesia. These diseases are transmitted through vectors such as mosquitoes, whose populations are heavily influenced by environmental factors, climate, and changes in human behavior (Kumar, Bauch, and Anand 2025) . The increasing in

FAQs

What machine learning models are used for classifying vector-borne diseases?

The study evaluates Random Forest, Logistic Regression, k-nearest Neighbors, Decision Tree, and XGBoost.

How does the study handle class imbalance in the dataset?

It applies oversampling methods including SMOTE and Random Oversampling to reduce bias toward majority classes.

Which models achieved the highest performance, and what does that imply?

Decision Tree and XGBoost achieved the highest accuracy at 99.2%, indicating that with proper data handling, models can improve early detection and support more timely public health decisions.

Aug 4, 2026

Upload by Aurelia