A COMPARATIVE STUDY USING MACHINE LEARNING FOR OBESITY PREDICTION
| dc.contributor.author | WINNIE TIONG RU PING | |
| dc.date.accessioned | 2026-04-27T23:50:00Z | |
| dc.date.issued | 2025 | |
| dc.description | Machine learning (ML) has proven effective in various fields, including healthcare, for tasks like obesity prediction, personalized medicine, and risk assessment. Cardiovascular diseases and diabetes are linked to obesity, a condition that is becoming a major worldwide health problem. Its increasing prevalence is driven by unhealthy diets, sedentary lifestyles, and genetic predispositions. Due to they do not consider many factors, including lifestyle and environmental impacts, traditional evaluation tools, such as the Body Mass Index (BMI) are insufficient. This study proposes an obesity prediction system using ML techniques to enhance accuracy in detecting and classifying obesity levels. The research employs the CRISP-DM methodology, encompassing stages such as business understanding, data understanding, data preparation, modeling, evaluation, and deployment. The dataset, obtained from Kaggle, includes diverse features related to basic information, eating habits, and physical conditions. Data transformation, feature scaling, and feature selection are examples of data preprocessing that make sure the dataset is appropriate for machine learning models. Five algorithms such as Decision Trees, Random Forest, Support Vector Machines (SVM), XGBoost, and Logistic Regression are used and evaluated for their performance in classifying obesity into categories aligned with WHO standards. Advanced methods like adjusting hyperparameters help to ensure the optimal performance of the models. Metrics such as accuracy, precision, recall, F1-score, and ROC-AUC are used to assess model performance and select the optimal method. Random Forest had the best performance overall. It gave the highest accuracy and balanced results across all metrics. The best model was used to build a web application. This app allows users to enter their data and get their obesity risk in real-time. The system helps people understand their health risks. It also supports doctors in giving faster assessments. | |
| dc.identifier.uri | https://scholarhub.unimas.my/handle/123456789/504 | |
| dc.language.iso | English | |
| dc.publisher | UNIVERSITI MALAYSIA SARAWAK | |
| dc.relation.ispartofseries | Faculty of Computer Science and Information Technology | |
| dc.title | A COMPARATIVE STUDY USING MACHINE LEARNING FOR OBESITY PREDICTION | |
| dc.type | Final Year Project |