Issue |
ITM Web Conf.
Volume 70, 2025
2024 2nd International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2024)
|
|
---|---|---|
Article Number | 04035 | |
Number of page(s) | 7 | |
Section | AI and Advanced Applications | |
DOI | https://doi.org/10.1051/itmconf/20257004035 | |
Published online | 23 January 2025 |
Comparisons of Machine Learning Models for Prediction of Susceptibility to Diabetes
Department of Mathematical Sciences, Carnegie Mellon University, 15213, Pittsburgh, United States
Corresponding author: yutianj@andrew.cmu.edu
Diabetes is a chronic disorder causing millions of people to suffer from severe complications such as heart attacks, kidney failures, and permanent vision loss. This study aims to find an optimal choice among the five selected models that perform the best on diabetes prediction, and thus provide valuable insights in early detection of diabetes. This study compares the predictive performance of machine learning models such as Random Forest (RF), Logistic Regression (LR), and Support Vector Machine (SVM). The study preprocessed the Pima Indians Diabetes (PID) dataset, and the models were trained on it before being assessed using four assessment criteria. According to the results, LR had the best accuracy of 0.76, with RF and SVM coming in second and third, respectively. Results showed that LR achieved the highest accuracy of 0.76, closely followed by RF and SVM. While SVM has the highest precision, it performs poorly on recall, limiting its overall performance on diabetes prediction. On the contrary, LR and RF achieved good results in the F-score, making them outperform the other models in terms of overall performance score in predicting diabetes.
© The Authors, published by EDP Sciences, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.