In Medical industry there are many diseases that makes a patient critical among them diabetes is one of the major disease that affect most of the people in early stage. Diabetes (or Diabetes Mellitus) is a group of metabolic diseases, chronic, in which there are high blood sugar levels and affects the body’s ability to use the energy found in food over a prolonged period. Researchers are finding effective methods for the prediction of diabetes. The main goal is to analysis the performance of various data mining techniques in the diabetes dataset for efficient extraction of valuable patterns. For doing so WEKA software was used as a mining tool for diagnosing the useful pattern. The Pima Indian diabetes dataset are used for the analysis. The dataset was applied in various classification algorithms to analysis the performance to identify an effective model that predict diabetes disease. In this, the analysis is done by applying attribute evaluator to enhance the accuracy then applying Naive Bayes, Bayes Net, J48 and Random Forest and the performance are compared. Through this study, Naive Bayes Algorithm provides better classification accuracy, when compared with classification algorithms like Bayes Net, J48 and Random Forest.

Diabetes, Health care, Naive Bayes, Bayes Net, J48 and Random Forest, WEKA

[1] Asma A Aljarullah. Decision tree discovery for the diagnosis type-2 diabetes. International conference on innovation in information technology. 2011; p. 303-7.
[2] Aiswarya Iyer, Jeyalatha S and Sumbaly Ronak. Diagnosis of diabetes using classification mining techniques. International Journal of Data Mining & Knowledge Management Process. 2015; 5:1-14. 2.
[3] “Bayes Net”, Wikipedia, Aug 2018.
[4] ChaitraliDangare, S. and SulabaApte,S.Improved study of disease prediction using data mining classification techmiques. Int.J.Comp.Appl., 2012,47(10):75-88.
[5] Global Diabetes Community, http://www.diabetes.co.uk/diabetes_care/blood-sugar-level-ranges.html
[6] Ianchao Han J, Juan C Rodriguze, Beheshti Mohsen. Diabetes Data Analysis and Prediction model discovery. Second International conference on future generation com- munication and networking. 2011; p. 96-9. 13.
[7] “J48”, Wikipedia, March 2018.
[8] K. Saravananathan and T. Velmurugan “Analyzing Diabetic Data using Classification Algorithms in Data Mining” Indian Journal ofScience and Technology, Vol 9 (43) | November 2016 | www.indjst.org
[9] Maniya Hardik, Mosin I Hasan, Komal P Patel. Comparative study of Naive Bayes Classifier and kNN for Tuberculosis. International Journal of Computer Applications. 2011; p. 22-6.
[10] “Naïve Bayes”, Wikipedia, March 2018.
[11] P.Yasodha and M. Kannan, "Analysis of a Population of Diabetic Patients Databases in WekaTool", International Journal of Scientific & Engineering Research, vol. 2, no. 5, 2011.
[12] “Random Forest”, Wikipedia, March 2018.
[13] Sankaranarayanan.S and Dr Pramananda Perumal.T, “Predictive Approach for Diabetes Mellitus Disease through Data Mining Technologies”, World Congress on Computing and Communication Technologies, 2014, pp. 231-233
[14] Sonu Kumari and Archana Singh, “A Data Mining Approach for the Diagnosis of Diabetes Mellitus”, Proceedings of71hlnternational Conference on Intelligent Systems and Control (ISCO 2013)
[15] Stutz J., P. Cheeseman. (1996) Bayesian classification (autoclass): Theory and results. In Advances in Knowledge Discovery and Data Mining. AAAI/MIT Press
[16] Uswa Ali Zia, Dr. Naeem Khan “Predicting Diabetes in Medical Datasets Using Machine Learning Techniques” International Journal of Scientific & Engineering Research Volume 8, Issue 5, May-2017, ISSN 2229-5518
[17] Velide Phani Kumar and Velide Lakshmi. A Data Mining Approach for Prediction and Treatment of diabetes Disease. International Journal of Science Inventions Today. 2014; 3:73-9.