Mortality Prediction of Diabetes and Parameter Analysis Based on Joint EDA and SVM
Jye-Lin Chien
2023
Abstract
In an era marked by a steadily increasing number of diabetes patients, diabetes has become a global concern. The World Health Organization (WHO) reported that between 2000 and 2019, the number of deaths linked to diabetes rose by 3%. As a result, the goal of this study is to look at the death rate among diabetics and give patients analytical insights to help them take quick preventive action against fatalities. First, Exploratory Data Analysis (EDA) techniques are utilized to visualize data and understand its features, structure, and relationships. Second, the Support Vector Machine (SVM) model is employed for classification tasks, aiming to find an effective hyperplane that separates these samples. Last, the obtained accuracy and highest cross-validation score can be used to analyze the performance of diabetes mortality rate analysis among different SVM models. After analysis and evaluation, the SVM linear kernel model has been identified as an effective classifier. Among the three SVM models with different kernels, the polynomial kernel exhibits the highest accuracy, while the linear kernel demonstrates the highest cross-validation score. Experimental findings underscore the substantial impact of the “diabetes pedigree function” on patient mortality rates.
DownloadPaper Citation
in Harvard Style
Chien J. (2023). Mortality Prediction of Diabetes and Parameter Analysis Based on Joint EDA and SVM. In Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML; ISBN 978-989-758-705-4, SciTePress, pages 481-485. DOI: 10.5220/0012799900003885
in Bibtex Style
@conference{daml23,
author={Jye-Lin Chien},
title={Mortality Prediction of Diabetes and Parameter Analysis Based on Joint EDA and SVM},
booktitle={Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML},
year={2023},
pages={481-485},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012799900003885},
isbn={978-989-758-705-4},
}
in EndNote Style
TY - CONF
JO - Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML
TI - Mortality Prediction of Diabetes and Parameter Analysis Based on Joint EDA and SVM
SN - 978-989-758-705-4
AU - Chien J.
PY - 2023
SP - 481
EP - 485
DO - 10.5220/0012799900003885
PB - SciTePress