Effectiveclassificationframework For Breast Tumorsusing Optimized Multi-Kernel Svm With Controlled Skewness

Document Type : Primary Research paper

Authors

Department Of Electronics And Communication Engineering, Bannari Amman Institute Of Technology, Sathyamangalam – 638401, India.

Abstract

The World İs So Advanced And Sophisticated Enough Nowadays. But, Cancer
Still Remains As A Deadly Disease İn Many Parts Of The World For All Living
Organisms. The İncidence Rate Of Cancer Among Humans Globally İs İncreasing
Steadily Day By Day.Among All Forms Of Tumors, Breast Cancer İs A Type Of İllness
That Plays A Major Role İn Disturbingthesurvivalrate Of Humans Globally.Thus, There
İs A Need To Predict Breast Cancer İn İts Earlier Stage. Thus, The Work İn This Paperis
Aided To Design A Robust Classification Model That İnvolves The Use Of A Randomized-
Parameter Optimizedmulti-Kernel Support Vector Machine (RPOMK-SVM) Classifier.
Before The Stage Of Classification, The Paper Analyzes The Nature Of İnput İn Order To
Obtain Promising Results. The Skewness Of The Feature Attributes İs Controlled And
Reformed Using Box-Cox Transform.For This Analysis And Evaluation, The
Paperemploysthe Breast Cancer Wisconsin (Diagnostic) Database,Which İs A Standard
Public Dataset. The Final Results Are Then Compared Against The Existing Algorithms.

Keywords