Title: Diagnosing obesity using classification-based machine learning models

Authors: Udeechee; T.V. Vijay Kumar; Aayush Goel

Addresses: School of Computer and Systems Sciences, Jawaharlal Nehru University, New Delhi, India ' School of Computer and Systems Sciences, Jawaharlal Nehru University, New Delhi, India ' Department of Electronics and Communication Engineering, Bharati Vidyapeeth's College of Engineering, New Delhi, India

Abstract: Obesity has been a major underlying risk for people with chronic diseases and, thus, needs to be diagnosed in the early stages. This requires clinical data related to potential obese individuals to be evaluated for diagnosing obesity levels in individuals. In this paper, classification based machine learning techniques have been applied on such clinical data to design obesity classification models, which would be capable of diagnosing whether an individual is obese or not. Classification techniques, including ensemble techniques, were used to design such obesity classification models. The performance of these obesity classification models was evaluated and compared on metrics such as, accuracy, precision, recall, F1-score and area under the receiver operating characteristic curve. Experimental results showed that the use of ensemble techniques improved the performance of the obesity classification models. Further, amongst the ensemble techniques, the boosting technique based obesity classification model performed the best.

Keywords: healthcare; disease; obesity; artificial intelligence; AI; machine learning; ML; classification techniques.

DOI: 10.1504/IJEH.2023.135784

International Journal of Electronic Healthcare, 2023 Vol.13 No.3, pp.247 - 273

Received: 25 May 2022
Accepted: 15 Jan 2023
Published online: 05 Jan 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article