Modified Linear Vector Quantization Technique for Classification of Heart Disease Data
Abstract
Early recognition of heart disease has importance in diagnosis. Neural networks are very efficient tools in field of classification. In this paper we have proposed a modified Linear vector quantization (LVQ) neural net for classification of heart disease data. In the Modified LVQ approach we introduced Mahalanobis distance calculation method which has improved the efficiency of classification. The classification is done on the basis of the various biomedical tests and statistics obtained from the tests. With the introduction of this new technique the diagnosis and classification of diseased person will be simpler and more accurate. We have also demonstrated a survey of three different classification techniques through experimental analysis.
Keywords: data mining; heart disease; Artificial Neural Network (ANN); Linear Vector Quantization (LVQ); Multilayer Perceptron (MLP); mahalanobis distance; Modified LVQ
Full Text:
PDFDOI: https://doi.org/10.26483/ijarcs.v3i4.1273
Refbacks
- There are currently no refbacks.
Copyright (c) 2016 International Journal of Advanced Research in Computer Science

