Diabetes prediction using Machine Learning algorithms and ontology

TL;DR

This paper reviews and compares machine learning and ontology-based techniques for diabetes prediction, highlighting the effectiveness of ontology classifiers and SVM in improving diagnostic accuracy.

Contribution

It provides a comparative analysis of various ML algorithms and introduces ontology-based methods for enhanced diabetes prediction accuracy.

Findings

Ontology classifiers achieved the highest accuracy.

SVM outperformed other algorithms in the study.

Ontology-based ML techniques showed promising results.

Abstract

Diabetes is one of the chronic diseases, which is increasing from year to year. The problems begin when diabetes is not detected at an early phase and diagnosed properly at the appropriate time. Different machine learning techniques, as well as ontology-based ML techniques, have recently played an important role in medical science by developing an automated system that can detect diabetes patients. This paper provides a comparative study and review of the most popular machine learning techniques and ontology-based Machine Learning classification. Various types of classification algorithms were considered namely: SVM, KNN, ANN, Naive Bayes, Logistic regression, and Decision Tree. The results are evaluated based on performance metrics like Recall, Accuracy, Precision, and F-Measure that are derived from the confusion matrix. The experimental results showed that the best accuracy goes for ontology classifiers and SVM.