DESIGN AND IMPLEMENTATION OF PREDICTIVE MODEL FOR PROGNOSIS OF DIABETES USING DATA MINING TECHNIQUES

Shuja Rashid Mirza, Sonu Mittal, Majid Zaman

Abstract


Purpose: The aim of this research is to design a predictive model using data mining tools and techniques that could be employed in prediction if diabetes, with the intension of enhancing the capability and efficiency of decision making.
Methods: The research was carried out on primary data that was collected from one of leading diagnostic centers in Srinagar (J&K). Data was preprocessed so as to remove inconsistencies’. Data mining techniques’ Naïve Bayes and support vector machine, K-nn were used for decision making. Data mining is a strong novel innovation for the extraction of concealed clairvoyant and significant facts from vast databases that can be utilized to increase thoughtful and novel insights. Utilizing superior data mining techniques to exhume vital knowledge is considered as a dissident way to enhance the quality and exactness of health services in order to improve healthcare service while bringing down the services cost and time.
Findings: The study demonstrated that model created using SVM outperformed Naïve Bayes and K-nn in predicting the disease

Keywords


Data mining; Decision support system; SVM; Naïve Bayes; K-nn; Diabetes; weka.

Full Text:

PDF


DOI: https://doi.org/10.26483/ijarcs.v9i2.5792

Refbacks

  • There are currently no refbacks.




Copyright (c) 2018 International Journal of Advanced Research in Computer Science