Comparison between ANN-Based Heart Stroke Classifiers Using Varied Folds Data Set Cross-Validation

Abstract

This paper presents the design and development of an artificial neural network (ANN) model for the classification of heart stroke disease. The novelty of this work is training multilayer perceptron (MLP) neural network architectures with back-propagation algorithm for multivariate large data sets. Subsequently, the performances of the ANN models are evaluated using database of heart stroke obtained from Cleveland Clinic Foundation Database with all attributes are numeric-valued. The accuracy of the designed ANN models are cross-validated using varied folds of data set comprising 303 instances extracted from different age groups. This study exhibits ANN-based prognosis for early detection of level of heart stroke with testing classification accuracy of 85.55 %. © Springer India 2015.