Authors: T. Satyanarayana Murthy
Addresses: Department of Information Technology, Chaitanya Bharathi Institute of Technology (CBIT), Hyderabad, India
Abstract: An unreasonable increase of glucose in blood results in diabetes. In recent times this problem is often seen in many people around the world. Having an efficient medical diagnosis of diabetic prevention is essential. In this context, healthcare professionals have come up with different solutions, but none of them have taken shape. Considering these facts, we have proposed an integrated diabetic prediction system with the inclusions of un-supervised K-means clustering and naive Bayes classification. In this context, random attribute selection is used as the initial centroid selection method for K-means clustering. Significant work is compared with traditional classification algorithms like naive Bayes, SVM, etc., in terms of accuracy along with additional performance parameters like sensitivity, specificity, precision, and F-measure. As a result, it is determined that the proposed algorithm with 99.42%, which is higher than any other recently proposed classification technique, is achieved.
Keywords: diabetes mellitus; metabolic; knowledge-based DSS; PIMA Indian Heritage; WEKA.
International Journal of Intelligent Enterprise, 2022 Vol.9 No.4, pp.408 - 421
Accepted: 17 Feb 2020
Published online: 25 Oct 2022 *