Please use this identifier to cite or link to this item: https://repositori.uma.ac.id/handle/123456789/29072
Title: Analisis Kinerja Metode Naive Bayes dalam Mengklasifikasikan Penyakit Diabetes
Other Titles: Performance Analysis of the Naive Bayes Method in Classifying Diabetes Disease
Authors: Nasution, Ilyas
metadata.dc.contributor.advisor: Hartono
Keywords: Analysis;Naïve Bayes;Naïve Bayes;Diabetes;Confusion Matrix
Issue Date: Sep-2025
Publisher: Universitas Medan Area
Series/Report no.: NPM;218160039
Abstract: Diabetes merupakan suatu penyakit kronis yang mempengaruhi jutaan orang di seluruh dunia dan menjadi tantangan besar bagi kesehatan masyarakat. Dalam beberapa tahun terakhir, metode machine learning telah banyak diterapkan di bidang medis untuk membantu klasifikasi dan diagnosis penyakit. Diantara berbagai algoritma klasifikasi, metode Naïve Bayes mendapat perhatian karena kesederhanaan, efisiensi, dan keefektifannya dalam menangani data probabilistik. Penelitian ini berfokus pada tiga varian Naïve Bayes, yaitu Gaussian Naïve Bayes, Bernoulli Naïve Bayes, dan Multinomial Naïve Bayes, yang masing-masing memiliki karakteristik unik yang menjadikannya cocok untuk berbagai jenis dataset dalam klasifikasi medis. Dataset yang digunakan dalam penelitian ini memiliki total 768 baris data dan jumlah 9 kolom. Pra-pemrosesan data meliputi pemeriksaan missing value, pembersihan data missing value, serta normalisasi data untuk menyamakan skala fitur numerik. Metode yang digunakan ialah Min-Max Scalling, yang mengubah nilai fitur menjadi berada dalam rentang 0 hingga 1. Dalam penelitian ini, dilakukan tiga skenario pembagian data dengan proporsi data yang berbeda yaitu 90:10, 80:20, dan 70:30. Evaluasi dilakukan menggunakan Confusion Matrix dengan metrik akurasi, presisi, recall, dan f1-score. Hasil dari pengujian menunjukkan bahwa metode Gaussian Naïve Bayes secara konsisten mendapatkan akurasi tertinggi dari tiga skenario pengujian. Pengujian pertama, kedua, dan ketiga metode Gaussian Naïve Bayes mendapatkan akurasi sebesar 80%, 82.28%, dan 76.27%. Dari ketiga pengujian tersebut, dapat disimpulkan bahwa Gaussian Naïve Bayes lebih optimal dalam mengklasifikasikan penyakit diabetes. Diabetes is a chronic disease that affects millions of people worldwide and poses a major challenge to public health. In recent years, machine learning methods have been widely applied in the medical field to assist in disease classification and diagnosis. Among various classification algorithms, the Naïve Bayes method has gained attention due to its simplicity, efficiency, and effectiveness in handling probabilistic data. This study focuses on three variants of Naïve Bayes, Gaussian Naïve Bayes, Bernoulli Naïve Bayes, and Multinomial Naïve Bayes, each of which has unique characteristics that make them suitable for different types of datasets in medical classification. The dataset used in this research consists of 768 rows and 9 columns. Data preprocessing includes checking for missing values, cleaning missing data, and normalizing data to standardize the scale of numerical features. The method used is Min-Max Scaling, which transforms feature values into a range between 0 and 1. In this study, three data-splitting scenarios were conducted with different proportions: 90:10, 80:20, and 70:30. Evaluation was performed using the Confusion Matrix with accuracy, precision, recall, and f1-score metrics. The experimental results show that the Gaussian Naïve Bayes method consistently achieved the highest accuracy across the three testing scenarios. In the first, second, and third experiments, Gaussian Naïve Bayes achieved accuracies of 80%, 82.28%, and 76.27%, respectively. Based on these results, it can be concluded that Gaussian Naïve Bayes is more optimal in classifying diabetes disease.
Description: 68 Halaman
URI: https://repositori.uma.ac.id/handle/123456789/29072
Appears in Collections:SP - Informatic Engineering

Files in This Item:
File Description SizeFormat 
218160039 - Ilyas Nasution - Fulltext.pdfCover, Abstract, Chapter I, II, III, V, Bibliography1.25 MBAdobe PDFView/Open
218160039 - Ilyas Nasution - Chapter IV.pdf
  Restricted Access
Chapter IV460.8 kBAdobe PDFView/Open Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.