IMPLEMENTASI PARTICLE SWARM OPTIMIZATION PADA ALGORITMA NAIVE BAYES, DECISION TREE C4.5, DAN RANDOM FOREST UNTUK KLASIFIKASI DIABETES MELITUS

MAULANA, REFFY (2025) IMPLEMENTASI PARTICLE SWARM OPTIMIZATION PADA ALGORITMA NAIVE BAYES, DECISION TREE C4.5, DAN RANDOM FOREST UNTUK KLASIFIKASI DIABETES MELITUS. S1 thesis, Universitas Mercu Buana Jakarta.

[img]
Preview
Text (HAL COVER)
01 COVER.pdf

Download (539kB) | Preview
[img] Text (BAB I)
02 BAB 1.pdf
Restricted to Registered users only

Download (59kB)
[img] Text (BAB II)
03 BAB 2.pdf
Restricted to Registered users only

Download (506kB)
[img] Text (BAB III)
04 BAB 3.pdf
Restricted to Registered users only

Download (57kB)
[img] Text (BAB IV)
05 BAB 4.pdf
Restricted to Registered users only

Download (963kB)
[img] Text (BAB V)
06 BAB 5.pdf
Restricted to Registered users only

Download (102kB)
[img] Text (DAFTRA PUSTAKA)
07 DAFTAR PUSTAKA.pdf
Restricted to Registered users only

Download (110kB)
[img] Text (LAMPIRAN)
08 LAMPIRAN.pdf
Restricted to Registered users only

Download (972kB)

Abstract

High blood sugar levels are a symptom of diabetes mellitus, a chronic metabolic condition brought on by abnormalities in the synthesis or usage of insulin. The World Health Organization (WHO) estimates that 1.5 million people die from diabetes each year, and that over 422 million people worldwide have the condition. Reliable techniques for early detection and categorization are crucial to supporting improved disease management because the rise in diabetes patients has greatly increased the global health burden. Using data from the 2023 BRFSS annual survey, this study compares and assesses how well the Naive Bayes, Decision Tree C4.5, and Random Forest algorithms perform in the classification of diabetes mellitus. Metrics including accuracy, precision, recall, and F1-score are used in the evaluation to gauge each algorithm's performance. This work also incorporates an optimization technique based on Particle Swarm Optimization (PSO) to enhance classification performance and reduce error. The research results show that Random Forest without PSO optimization achieved the highest accuracy of 95.21%. Although PSO successfully improved Naïve Bayes accuracy from 80.74% to 82.08% and C4.5 from 91.24% to 91.57%, Random Forest experienced a slight decrease in accuracy after optimization to 94.60%, indicating that the model was already highly optimal. Further analysis of feature importance revealed that attributes such as GenHealth, BMI, and Age were consistently the most significant predictors for Diabetes Mellitus classification. It is anticipated that this study will be used as a guide for the creation of precise and effective data-driven medical applications, especially those that aid in the early detection of diabetes. As a result, this work advances the creation of technology solutions based on machine learning to address issues in the field of global health. Kata kunci: Diabetes, Classification, Naive Bayes, Decision Tree C4.5, Random Forest, Particle Swarm Optimization Kadar gula darah tinggi merupakan gejala Diabetes Melitus, kondisi metabolik kronis yang disebabkan oleh kelainan dalam sintesis atau penggunaan insulin. Organisasi Kesehatan Dunia (WHO) memperkirakan bahwa 1,5 juta orang meninggal karena diabetes setiap tahun, dan lebih dari 422 juta orang di seluruh dunia menderita kondisi tersebut. Teknik yang andal untuk deteksi dini dan kategorisasi sangat penting untuk mendukung peningkatan manajemen penyakit karena peningkatan jumlah pasien diabetes telah sangat meningkatkan beban kesehatan global. Menggunakan data dari survei tahunan BRFSS 2023, studi ini membandingkan dan menilai seberapa baik kinerja algoritma Naive Bayes, Decision Tree C4.5, dan Random Forest dalam klasifikasi diabetes melitus. Metrik termasuk accuracy, precision, recall, dan F1-score digunakan dalam evaluasi untuk mengukur kinerja setiap algoritma. Pekerjaan ini juga menggabungkan teknik optimasi Particle Swarm Optimization (PSO) untuk meningkatkan kinerja klasifikasi dan mengurangi kesalahan. Hasil penelitian menunjukkan bahwa Random Forest tanpa optimasi PSO mencapai akurasi tertinggi sebesar 95.21%. Meskipun PSO berhasil meningkatkan akurasi Naïve Bayes dari 80.74% menjadi 82.08% dan C4.5 dari 91.24% menjadi 91.57%, Random Forest mengalami sedikit penurunan akurasi setelah dioptimasi menjadi 94.60%, mengindikasikan bahwa model tersebut sudah sangat optimal. Analisis fitur penting (feature importance) lebih lanjut mengungkapkan bahwa atribut seperti GenHealth, BMI, dan Age secara konsisten menjadi prediktor paling signifikan untuk klasifikasi Diabetes Melitus. Penelitian ini diharapkan dapat digunakan sebagai panduan untuk pembuatan aplikasi medis berbasis data yang akurat dan efektif, terutama yang membantu dalam deteksi dini diabetes. Hasilnya, karya ini memajukan penciptaan solusi teknologi berdasarkan machine learning untuk mengatasi berbagai masalah di bidang kesehatan global. Kata kunci: Diabetes, Klasifikasi, Naive Bayes, Decision Tree C4.5, Random Forest, Particle Swarm Optimization

Item Type: Thesis (S1)
Call Number CD: FIK/INFO. 25 099
NIM/NIDN Creators: 41521010103
Uncontrolled Keywords: Diabetes, Klasifikasi, Naive Bayes, Decision Tree C4.5, Random Forest, Particle Swarm Optimization
Subjects: 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 004 Data Processing, Computer Science/Pemrosesan Data, Ilmu Komputer, Teknik Informatika
500 Natural Science and Mathematics/Ilmu-ilmu Alam dan Matematika > 510 Mathematics/Matematika > 518 Numerical Analysis/Analisis Numerik, Analisa Numerik > 518.1 Algorithms/Algoritma
600 Technology/Teknologi > 630 Agriculture and Related Technologies/Pertanian dan Teknologi Terkait
Divisions: Fakultas Ilmu Komputer > Informatika
Depositing User: khalimah
Date Deposited: 02 Aug 2025 03:31
Last Modified: 02 Aug 2025 03:31
URI: http://repository.mercubuana.ac.id/id/eprint/96469

Actions (login required)

View Item View Item