Abstract: Diabetes is a chronic disease that spread over the past decades in abundance. It is a metabolic disease that may affect the entire body. Diabetes is classified are three types, which are type 1 diabetes (T1D), type 2 diabetes (T2D), and gestational diabetes (GD), where each type has specific causes. This research study aims to find out the most common behaviours that lead to diabetes and measure the relationship between human biometrics and the likelihood of behaving T2D.The study aimed to develop a machine learning prediction model by investigating five machine learning algorithms which are Support Vector Machine, Logistic Regression, K-Nearest Neighbour, Decision Tree, and Random Forest. This model was developed by Python using google colab, Random Forest algorithm outperformed in perform highly accurate behavioural prediction with 98% compared with other algorithms. The outcome from this research study would assist the medical practice and medical community with a tool that can early predict T2D.
Keywords: Behaviours, Diabetes mellitus, Machine Learning, Prediction, Type 2 diabetes.
Works Cited:
Samah Alzahrani "An Intelligent Model for early prediction of Type 2 diabetes likelihood using human behaviors and biometrics among adults in Saudi Arabia", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 12, no. 6, pp. 328-338, 2023. Crossref https://doi.org/10.17148/IJARCCE.2023.12657
|
DOI:
10.17148/IJARCCE.2023.12657
