Using Artificial Intelligence Techniques for the Analysis of Obesity Status According to the Individuals\' Social and Physical Activities

Authors : Nigmet Koklu, Süleyman Alpaslan Sulak

Pages : 217-239

Doi:10.33484/sinopfbd.1445215

View : 56 | Download : 66

Publication Date : 2024-06-29

Article Type : Research

Abstract :Obesity is a serious and chronic disease with genetic and environmental interactions. It is defined as an excessive amount of fat tissue in the body that is harmful to health. The main risk factors for obesity include social, psychological, and eating habits. Obesity is a significant health problem for all age groups in the world. Currently, more than 2 billion people worldwide are obese or overweight. Research has shown that obesity can be prevented. In this study, artificial intelligence methods were used to identify individuals at risk of obesity. An online survey was conducted on 1610 individuals to create the obesity dataset. To analyze the survey data, four commonly used artificial intelligence methods in literature, namely Artificial Neural Network, K Nearest Neighbors, Random Forest and Support Vector Machine, were employed after pre-processing. As a result of this analysis, obesity classes were predicted correctly with success rates of 74.96%, 74.03%, 74.03% and 87.82%, respectively. Random Forest was the most successful artificial intelligence method for this dataset and accurately classified obesity with a success rate of 87.82%.
Keywords : Obezite veri seti, yapay zeka yöntemleri, yapay sinir ağı, destek vektör makinesi, k- en yakın komşu, rastgele orman