A REVIEW ON OBESITY PREDICTION USING MACHINE LEARNING TECHNIQUES - Research paper

Aug 3, 2026

Obesity has become a major global health crisis, creating long-term risks and increasing the burden on both healthcare systems and patients. Timely identification supports preventive action by enabling earlier intervention and reducing the chance that underlying causes or symptoms worsen. Manual review of medical history can be slow, error-prone, and costly, motivating the development of automated predictive models. This review examines machine learning approaches for obesity prediction and prevention, focusing on multiple classifiers and reporting performance outcomes, with Random Forest achieving the highest accuracy.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages4

Reading Time10 min

AccessFree to download

Abstract

Introduction

Review of Literature

20(2): S2: 25-28, 2025 [www.thebioscan.com](www.thebioscan.com) A REVIEW ON OBESITY PREDICTION USING MACHINE LEARNING TECHNIQUES Dr. Kolluru Venkata Nagendra1, Dr. Praveen B M2 PDF Research Scholar 1, Research Guide2 Department of Computer Science and Engineering1 2, Srinivasa University, Mangalore, Karnataka State, India.1 2 DOI: [https://doi.org/10.63001/tbs.2025.v20.i02.S2.pp25-28](https://doi.org/10.63001/tbs.2025.v20.i02.S2.pp25-28) KEYWORDS Detection of Obesity, Machine Learning, Logistic Regression, Decision Tree, KNN, SVM. Received on: 12-02-2025 Accepted on: 10-03-2025 Published on: 08-04-2025 ABSTRACT At present, protecting the community is crucial for addressing health issues, which can be done through medical research. Obesity has emerged as a global health crisis, posing a significant risk to the future. It ranks as one of the most prevalent health issues worldwide. Timely identification of a disease can assist both healthcare professionals and patients in taking action to reduce, if not completely eliminate, the underlying cause or in preventing the symptoms of the disease from worsening. Reviewing a patient's medical history is a common approach to diagnosing a disease; however, this process can be quite time-consuming when done manually and is often susceptible to errors and high costs. Thus, there is a compulsory to scientifically create a predictive model for the development of diseases using automated methods in today's world. This research highlights the capabilities of machine learning in tackling public health issues, laying the groundwork for future studies aimed at improving obesity prediction and prevention methods. Four machine learning algorithms were engaged: Random Forest, Decision Tree, K-Nearest Neighbor, and Support Vector Machine. The results have been encouraging, with the Random Forest classifier got the highest accuracy at 96.93% among all. INTRODUCTION Globally, Obesity is one of the most frequent health issue linked to numerous illnesses, risks, and even mortality. It represents a significant health challenge on a worldwide scale, emerging as a potential danger for the future. As noted in [1] Mendoza, P.(2019), obesity has become a widespread and evolving global epidemic that has surged since the 1980s, raising serious health issues among adults, adolescents, and children alike. Furthermore, Eduado, D. , Fabio, E. , pointed out that the challenges associated with obesity are escalating swiftly, prompting new research that addresses obesity detection in both the young and the elderly. According to this understanding, the WHO. (2021, June 23) . Obesity Related Diseases [2] tells obesity as a rare or excessive collection of fat that can adversely affect health. Many individuals over the age of 16 are experiencing changes in their weight. This excess weight can result from a high consumption of starchy foods rich in fat, as well as a deficiency in physical activity. As noted in Guterrez, H. M. (2010) . [3], obesity is a widespread health issue globally, impacting teenagers, children, and adults alike. Obesity can be regarded as a multifaceted disease influenced by various factors, characterized by symptoms such as uncontrolled weight gain, primarily due to excessive consumption of energydense and fatty foods. Hernandez, J. (2011) [4] Research indicates that biological factors, including genetic predisposition, play a significant role in the development of different types of obesity, such as syndromic, monogenic, polygenic and leptin. Additionally, other risk factors, including dietary habits, social influences and psychological aspects, have been identified. Globally, [5] represents a significant health challenge, as it is associated with chronic conditions such as cardiovascular diseases, diabetes, and cancer. Early detection of obesity has become a focal point in health initiatives. Nevertheless, extensive research over the years has revealed that

FAQs

Why is obesity prediction important in public health?

Obesity is linked to multiple chronic diseases and increases health risks, making early identification essential for timely prevention and intervention.

What machine learning algorithms are discussed for obesity prediction?

The review engages Random Forest, Decision Tree, K-Nearest Neighbor, and Support Vector Machine, comparing their predictive performance.

Which algorithm achieved the highest accuracy and what was the value?

Random Forest produced the highest accuracy at 96.93% among the evaluated algorithms.

Aug 3, 2026

Upload by Clementine