PERFORMANCE COMPARASION OF K-NEAREST NEIGHBOR, NAIVE BAYES, AND RANDOM FOREST ALGORITMS IN OBESITY PREDICTION

Andani, Mia and Triloka, Joko (2025) PERFORMANCE COMPARASION OF K-NEAREST NEIGHBOR, NAIVE BAYES, AND RANDOM FOREST ALGORITMS IN OBESITY PREDICTION. PERFORMANCE COMPARASION OF K-NEAREST NEIGHBOR, NAIVE BAYES, AND RANDOM FOREST ALGORITMS IN OBESITY PREDICTION, 9 (3). pp. 502-510. ISSN 2541-2019

[img] Text
COVER MIA ANDANI.pdf
Download (46kB)
[img] Text
pengesahan publikasi.pdf
Download (130kB)
[img] Text
pernyataan keaslian laporan.pdf
Download (140kB)
[img] Text
persetujuan publikasi.pdf
Download (128kB)
[img] Text
14478-Article Text-21581-1-10-20250211.pdf
Download (336kB)
[img] Text
Cek Turnitin.pdf
Download (2MB)
[img] Text
daftar isi.pdf
Download (94kB)
[img] Text
LoA.pdf
Download (174kB)

Abstract

-Abstract: Obesity is a growing global health issue that has serious impacts on both physical and mental health. According to the World Health Organization (WHO), over 1.9 billion adults worldwide are overweight, with more than 650 million of them categorized as obese. Early detection of obesity is a crucial step to prevent further complications, however, traditional methods such as Body Mass Index (BMI) have limitations in distinguishing between muscle mass and body fat. This study aims to predict an individual's obesity status based on specific attributes using the K-Nearest Neighbor (K-NN), Naive Bayes, and Random Forest algorithms. The dataset used was sourced from the Kaggle platform, containing 2,111 records and 16 attributes, including gender, age, weight, height, frequency of high-calorie foods, physical activity, and water and vegetable consumption patterns. The research process follows the data mining stages, including business understanding, data understanding, data preparation, modeling, evaluation, and documentation. Experiments were conducted using RapidMiner with a 10-fold cross-validation technique to assess the overall model performance. The results show that the Random Forest algorithm outperforms K-NN and Naive Bayes in predicting obesity status. Model evaluation using metrics such as accuracy, precision, recall, and F1-score demonstrates significant results in distinguishing obesity categories. It is hoped that this research can contribute to developing machine learning-based health prediction systems that can support decision-making in the prevention and management of obesity. Keywords: Obesity, Data Classification, K-Nearest Neighbor, Naïve Bayes, Data Mining.

Item Type: Article
Subjects: Ilmu Komputer
eSkripsi
Divisions: Pasca Sarjana > Magister Teknik Informatika
Depositing User: Mia Andani
Date Deposited: 17 Jun 2025 04:01
Last Modified: 17 Jun 2025 04:01
URI: http://repo.darmajaya.ac.id/id/eprint/20433

Actions (login required)

View Item View Item