Diabetes Diagnosis from Big Data using Fuzzy-Neural Chaotic Tree

Document Type : Original Article


1 Department of Information Technology Sabzevar Islamic Azad University Sabzevar, Iran

2 Department of Computer Engineering, Sabzevar Branch, Islamic Azad University Sabzevar, Iran


Today, diabetes is recognised as a significant global health issue. Worldwide statistics indicate an increasing prevalence of the disease, thereby presenting a challenging problem for modern medicine.  To address this issue, computer science has presented various methods for the diagnosis and prediction of diabetes. However, there remain unresolved issues and errors that researchers endeavour to rectify. Data mining is employed as a technical means of identifying and extracting fresh knowledge from data. This research presents a new technique for categorising diabetic data, which involves three stages. Firstly, preprocessing is carried out, in which data normalisation procedures are performed. This is followed by attribute extraction and selection. Finally, classification takes place using data mining principles. The obtained classification outcomes can be employed to forecast diabetes in varied individuals. The results are evaluated using certain standards, such as sensitivity, specificity, and accuracy. Our suggested methodology, which combines chaotic fuzzy-neural with K-means tree, is more appropriate than previous techniques, as confirmed by the outcomes.