An Intelligent Data Mining Model for Diabetic Patients Data Using Hybrid Adaptive Neuro- Fuzzy Inference System
S.Abinesh1, G. Prabakaran2, R. Arunkumar3
1S. Abinesh, Research Scholar, Department of Computer Science and Engineering, Annamalai University, Tamil Nadu, India.
2Dr. G. Prabakaran, Associate Professor, Department of Computer Science and Engineering, Annamalai University, Tamil Nadu, India.
3Dr. R. Arunkumar, Associate Professor, Department of Computer Science and Engineering, Annamalai University, Tamil Nadu, India.
Manuscript received on July 20, 2019. | Revised Manuscript received on August 10, 2019. | Manuscript published on August 30, 2019. | PP: 1498-1503 | Volume-8 Issue-6, August 2019. | Retrieval Number: F8136088619/2019©BEIESP | DOI: 10.35940/ijeat.F8136.088619
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Proper diagnosis of diabetic based on the patient’s medical analysis results is an important factor. Data mining helps in analyzing such data includes complex meaningful terms to diagnosis and supports the patients to take remedy action based on the accurate results. The proposed model is a data mining model for analyzing diabetic patient’s data using sugeno type adaptive neuro fuzzy inference system with principle component analysis as a hybrid system. The experimental model validated through 200 different data obtained from health clinic with 25 different attributes. The proposed model classifies the data with accuracy of 94.6% where as conventional rough set and k means clustering model produces less classification accuracy of 74.5% and 77.6%.
Keywords: Diabetic Data, ANFIS, PCA, and Data mining.