K-Fold Cross Validation and Classification Accuracy of PIMA Indian Diabetes Data Set using Higher Order Neural Network and PCA
Raj Anand1, Vishnu Pratap Singh Kirar2, Kavita Burse3
1Raj Anand, Department of Computer Science, Oriental College of Technology, Bhopal, India.
2Vishnu Pratap Singh Kirar, Department of Electronics & Communication, Truba Institute of Engineering & Information Technology, Bhopal, India.
3Dr. Kavita Burse, Department of Electronics & Communication, Oriental College of Technology, Bhopal, India.
Manuscript received on January 01, 2013. | Revised Manuscript received on January 02, 2013. | Manuscript published on January 05, 2013. | PP: 436-438 | Volume-2, Issue-6, January 2013. | Retrieval Number: F1212112612/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: 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: Neural network techniques have been successfully applied for diagnosis of Type II diabetes. We propose a K-Fold cross validation method for classification of PIMA Indian diabetes data set. The classification accuracy is computed with PCA preprocessing and higher order neural network. The problem of missing data in the analysis and decision making process is handled through PCA. PCA also scales the data in the same range of values.
Keywords: Type II diabetes, Pima Indian data set, higher order neural networks, data pre-processing, K cross validation, PCA.