OPTIMASI PREDIKSI RISIKO STROKE MENGGUNAKAN ARTIFICIAL NEURAL NETWORK DENGAN SELEKSI FITUR BERBASIS BINARY PARTICLE SWARM OPTIMIZATION

ANUGERAH, NAUFAL RAHFI (2026) OPTIMASI PREDIKSI RISIKO STROKE MENGGUNAKAN ARTIFICIAL NEURAL NETWORK DENGAN SELEKSI FITUR BERBASIS BINARY PARTICLE SWARM OPTIMIZATION. S1 thesis, Universitas Mercu Buana Jakarta.

Preview	Text (HAL COVER) Cover.pdf Download (623kB) | Preview
	Text (BAB I) Bab 1.pdf Restricted to Registered users only Download (196kB)
	Text (BAB II) Bab 2.pdf Restricted to Registered users only Download (403kB)
	Text (BAB III) Bab 3.pdf Restricted to Registered users only Download (353kB)
	Text (BAB IV) Bab 4.pdf Restricted to Registered users only Download (386kB)
	Text (BAB V) Bab 5.pdf Restricted to Registered users only Download (183kB)
	Text (DAFTAR PUSTAKA) Daftar Pustaka.pdf Restricted to Registered users only Download (213kB)
	Text (LAMPIRAN) Lampiran.pdf Restricted to Registered users only Download (750kB)

Abstract

Stroke remains a major global cause of mortality, and early prediction is vital for preventive healthcare. This study introduces an Artificial Neural Network optimized by Binary Particle Swarm Optimization (ANN-BPSO) combined with Synthetic Minority Over-sampling Technique for Nominal and Continuous (SMOTE-NC) to address class imbalance in mixed-type medical data. The framework adopts a leakage-free evaluation, ensuring oversampling is applied only to the training set while maintaining the authenticity of the test data. Experimental results indicate that the proposed model produces more balanced and realistic outcomes than previous studies. Rather than yielding inflated metrics, the model reflects true predictive behaviour consistent with real data distributions. By integrating feature optimization with controlled resampling, the approach enhances stability, discrimination, and clinical reliability in stroke-risk prediction. Keywords: Stroke Risk Prediction, Artificial Neural Network, Binary Particle Swarm Optimization, Machine Learning, Data Leakage Prevention. Stroke tetap menjadi salah satu penyebab utama kematian secara global, dan prediksi dini sangat penting untuk layanan kesehatan preventif. Studi ini memperkenalkan Artificial Neural Network yang dioptimalkan dengan Binary Particle Swarm Optimization (ANN-BPSO) yang dipadukan dengan Synthetic Minority Over-sampling Technique for Nominal and Continuous (SMOTE-NC) untuk menangani ketidakseimbangan kelas pada data medis bertipe campuran. Kerangka kerja ini menggunakan evaluasi bebas kebocoran, memastikan bahwa oversampling hanya diterapkan pada data pelatihan sambil menjaga keaslian data pengujian. Hasil eksperimen menunjukkan bahwa model yang diusulkan menghasilkan keluaran yang lebih seimbang dan realistis dibandingkan penelitian sebelumnya. Alih-alih menghasilkan metrik yang berlebihan, model ini mencerminkan perilaku prediktif yang sesuai dengan distribusi data sebenarnya. Dengan mengintegrasikan optimasi fitur dan resampling terkontrol, pendekatan ini meningkatkan stabilitas, kemampuan diskriminasi, dan keandalan klinis dalam prediksi risiko stroke. Kata Kunci: Prediksi Risiko Stroke, Artificial Neural Network, Binary Particle Swarm Optimization, Pembelajaran Mesin, Anti Kebocoran Data.

Item Type:	Thesis (S1)
NIM/NIDN Creators:	41822010038
Uncontrolled Keywords:	Prediksi Risiko Stroke, Artificial Neural Network, Binary Particle Swarm Optimization, Pembelajaran Mesin, Anti Kebocoran Data.
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 > 006 Special Computer Methods/Metode Komputer Tertentu > 006.3 Artificial Intelligence/Kecerdasan Buatan 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 > 006 Special Computer Methods/Metode Komputer Tertentu > 006.3 Artificial Intelligence/Kecerdasan Buatan > 006.31 Machine Learning/Pembelajaran Mesin
Divisions:	Fakultas Ilmu Komputer > Sistem Informasi
Depositing User:	khalimah
Date Deposited:	14 Feb 2026 02:51
Last Modified:	14 Feb 2026 02:51
URI:	http://repository.mercubuana.ac.id/id/eprint/100939

Actions (login required)

View Item