A Machine Learning in Binary and Multiclassification Results on Imbalanced Heart Disease Data Stream

Abstract

In medical filed, predicting the occurrence of heart diseases is a significant piece of work. Millions of healthcare-related complexities that have remained unsolved up until now can be greatly simplified with the help of machine learning. The proposed study is concerned with the cardiac disease diagnosis decision support system. An OpenML repository data stream with 1 million instances of heart disease and 14 features is used for this study. After applying to preprocess and feature engineering techniques, machine learning approaches like random forest, decision trees, gradient boosted trees, linear support vector classifier, logistic regression, one-vs-rest, and multilayer perceptron are used to perform binary and multiclassification on the data stream. When combined with the Max Abs Scaler technique, the multilayer perceptron performed satisfactorily in both binary (Accuracy 94.8%) and multiclassification (accuracy 88.2%). Compared to the other binary classification algorithms, the GBT delivered the right outcome (accuracy of 95.8%). Multilayer perceptrons, however, did well in multiple classifications. Techniques such as oversampling and undersampling have a negative impact on disease prediction. Machine learning methods like multilayer perceptrons and ensembles can be helpful for diagnosing cardiac conditions. For this kind of unbalanced data stream, sampling techniques like oversampling and undersampling are not practical.

An Ensemble Feature Optimization for an Effective Heart Disease Prediction Model

“…In this section the heart disease prediction using proposed EFO and other existing ML-based heart disease prediction method are studied [15,17]. The UCI repository Cleveland dataset from [27,28] is used for performance analysis.…”

“…The UCI repository Cleveland dataset from [27,28] is used for performance analysis. The selection of dataset is based on comparison paper [15,17]. The machine learning model for performing heart disease prediction is implemented using python 3 framework.…”

“…In [13,14], they failed to establish which feature is important for the prediction of the heart disease using standard cross validation algorithm. The proposed model in [15][16][17] exhibited poor classification accuracy when the data is imbalanced in nature. Further, the proposed models in [21][22][23][24][25] mainly focussed on the binary classification and did not address the multi-classification issue.…”

“…In [17] is focused with design of the cardiac disease diagnosis decision support system. This research makes advantage of a stream of data from the OpenML repository that contains 1,000,000 records of cardiac illness and 14 characteristics [18][19][20].…”

“…Recently, ensemble technique [12] combining several classifiers together have been presented for improving the accuracies of heart disease prediction. However, these models [13,14] failed to establish which feature is important in predicting heart disease using standard cross validation algorithm; Further, existing heart disease prediction model [15,16] exhibits very poor classification accuracy when data is imbalanced [17] in nature. This paper presents an ensemble feature optimization (EFO) for effective heart disease (EDH) predication model.…”