People in today's world get affected by many diseases that do not have a complete cure. The development of one disease may lead to various other complications. One such disease is Type-2 Diabetes. It is a global health problem. This is the most common type of diabetes usually developed at the age of 40 and older. This increases the risk factors like kidney failure, heart disease, blindness, nerve damage and blood vessel damage. It is predicted from the characteristics of the patients. A Hybrid Prediction Model (HPM) has been developed using k-means clustering and C4. 5 classifier. In that model, the dataset is initially cleaned and then Z-score normalization is applied on this dataset. A pattern is extracted from this using clustering. A model was built on this extracted pattern using the c4. 5 classifier. This produced an accuracy of 92. 38%. This classification accuracy can be improved by using Neural network. This improved model separates the dataset into either one of the two groups. This model has revealed an accuracy of 97. 93%. This earlier detection will help the physicians to reduce the probability of getting that disease.

1. Jianchao Han, Juan C. Rodriguze, Mohsen Beheshti, "Diabetes Data Analysis and Prediction Model Discovery Using Rapid miner", IEEE, Second International Conference on Future Generation Communication and Networking,pp. 96-99,2008.
2. Asma A. AlJarullah, "Decision tree discovery for the diagnosis of type-2 Diabetes", IEEE, 2011 Internaional Conference on innovations in information technology, pp. 303-307, 2011.
3. B. M. Patil, R. C. Joshi, Durga Toshniwal, "Hybrid Prediction Model for Type-2 Diabetic patients", Expert Systems with Applications, Science direct, pp. 8102-8108,2010.
4. "Diabetes Mellitus: a guide to patient care", Williams L and Wilkins, Philadelphia, 2007.
5. http://www. effectsofdiabetes. org/
6. http://archive. ics. uci. edu/ml/machine-learning-databases/pima-indians-diabetes/pima-indians-diabetes. data
7. C. Yue, L. Xin, X. Kewen and S Chang, "An Intelligent Diagnosis to Type 2 Diabetes Based on QPSO algorithm and WLS-SVM", IEEE, International Symposium on Intelligent Information Technology Application Workshops,2008.
8. E. G. Yildirim, A. Karahoca and T. Ucar, 'Dosage planning for diabetes patients using data mining methods', Science Direct , Procedia Computer science, pp. 1374-1380,2010.
9. S. M Nuwangi, C. R. Oruthotaarachchi, J. M. P. P Tilakaratna and H. A. Caldera, 'Usage of Association rules and Classification Techniques in Knowledge Extraction of Diabetes' IEEE, pp. 372-377,2010.

Classification Clustering Z-score Normalization