Title :
Efficient classification and analysis of ischemic heart disease using proximal support vector machines based decision trees
Author_Institution :
Amrita Inst. of Manage., Amrita Univ., Coimbatore, India
Abstract :
Ischemic heart disease (IHD) is one of the toughest challenges to doctors in-making right decisions due to its skimpy symptoms and complexity. We have analyzed IHD data from 65 patients to provide an aid for decision-making. Decision trees give potent structural information about the data and thereby serve as a powerful data mining tool. Support vector machines serve as excellent classifiers and predictors and can do so with high accuracy. Our tree based classifier uses non-linear proximal support vector machines (PSVM). The accuracy is very high (100% for training data) and the tree is small and precise.
Keywords :
data mining; decision support systems; decision trees; medical signal processing; patient diagnosis; signal classification; support vector machines; IHD; PSVM; classifiers; data mining tool; decision-making aids; differential diagnosis; heart disease analysis; ischemic heart disease classification; nonlinear proximal support vector machines; predictors; proximal support vector machines based decision trees; Cardiac disease; Cardiovascular diseases; Classification tree analysis; Computer networks; Data mining; Decision trees; Ischemic pain; Magnetic heads; Support vector machine classification; Support vector machines;
Conference_Titel :
TENCON 2003. Conference on Convergent Technologies for the Asia-Pacific Region
Print_ISBN :
0-7803-8162-9
DOI :
10.1109/TENCON.2003.1273317
