ABSTRACT
Obesity denotes the unusual or extreme fat increase in the human body that can be quite dangerous for individuals’ health. It is associated with adverse health conditions including lung problems, heart diseases, type 2 diabetes, strokes, and cancers. In addition to physical measures, features related to individuals’ food habits and lifestyles are considered to estimate obesity levels. The objective of this paper is to use machine learning techniques to detect individuals’ obesity levels. Specifically, the research study applied various machine learning-based classification techniques such as Logistic Regression, Support Vector Machine, Random Forest, Multilayer Perceptron, and Naive Bayes to detect obesity. The research found Logistic Regression to outperform all the other techniques. In other words, Logistic Regression could be a promising method for use in the medical field such as obesity detection. Future works can broaden the scope of this research by studying the role of machine learning techniques in the identification of other adverse health problems.
