Hybrid data mining model for the classification and prediction of medical datasets
by S. Raghavendra; M. Indiramma
International Journal of Knowledge Engineering and Soft Data Paradigms (IJKESDP), Vol. 5, No. 3/4, 2016

Abstract: Prediction and classification of medical datasets help in reducing the number of diagnostic procedure to identify the diseases there by providing economical solutions for healthcare systems and medical diagnosis software system. Data pre-processing plays a vital role in preparing effective and efficient data for data mining. Feature selection helps in providing important attributes for building comprehensive predictive models. In many existing work on classification and prediction of medical datasets, either full set of attributes or reduced set of attributes are used to find the classification accuracy. But the classification accuracy achieved from these is not satisfactory. Hence, in this paper we propose hybrid data mining predictive model for attribute selection based on entropy evaluation, mean evaluation, and threshold evaluation. In this prediction model we use feature subset selection methods (FSM) like forward selection and backward elimination method to select the best subset of attributes to reduce training time and to improve performance of the system.

Online publication date: Sat, 17-Jun-2017

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Knowledge Engineering and Soft Data Paradigms (IJKESDP):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com