CLINICAL DATABASE ANALYSIS USING DMDT BASED PREDICTIVE MODELLING
Abstract
vioft2nntf2t|tblJournal|Abstract_paper|0xf4ffaf710f0000006f3b020001000500
In recent years, predictive data mining techniques play a vital role in the field of medical informatics. These techniques help the medical practitioners in predicting various classes which is useful in prediction treatment. One of such major difficulty is prediction of survival rate in breast cancer patients. Breast cancer is a common disease these days and fighting against it is a tough battle for both the surgeons and the patients. To predict the survivability rate in breast cancer patients which helps the medical practitioner to select the type of treatment a predictive data mining technique called Diversified Multiple Decision Tree (DMDT) classification is used. Additionally, to avoid difficulties from the outlier and skewed data, it is also proposed to perform the improvement of training space by outlier filtering and over sampling. As a result, this novel approach gives the survivability rate of the cancer patients based on which the medical practitioners can choose the type of treatment.

Authors
Srilakshmi Indrasenan, Sathiyabhama Balasubramaniam
Sona College of Technology, India

Keywords
Predictive Data Mining, Breast Cancer, Survivability Rate, Outlier Filtering, Diversified Multiple Decision Tree
Yearly Full Views
JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember
000000000102
Published By :
ICTACT
Published In :
ICTACT Journal on Soft Computing
( Volume: 3 , Issue: 3 , Pages: 549-554 )
Date of Publication :
April 2013
Page Views :
222
Full Text Views :
3

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.