Machine Learning-based Classification of Diabetes Disease: A Case Study with Orange Data Mining

Timely diagnosis plays a pivotal role in saving lives, necessitating the integration of artificial intelligence with medical data to achieve real-time accuracy. In this research, a variety of Machine Learning algorithms, including Support Vector Machines, Random Forest, Naive Bayes, Logistic Regression, K-nearest neighbors, and Decision Tree, were applied to detect diabetes using the PIMA Indians diabetes dataset comprising 9 attributes and 768 cases. The primary objective of this study is to compare the performance of these classification algorithms. The evaluation was based on key metrics such as Accuracy, Precision, F1 score, and Recall. The machine learning models exhibited commendable accuracy, particularly with the Random Forest algorithm achieving 97% accuracy, the Decision Tree reaching 95% accuracy, the Support Vector Machines attaining 86% accuracy, and the K-nearest neighbors demonstrating 85% accuracy.