Arrythmia prediction from high dimensional electrocardiogram’s data corpus using ensemble classification

In clinical practice, software aided arrhythmia diagnosis from electrocardiographic signals is critical, and it has the capability to minimize mortality induced by untrained clinicians. Furthermore, computer-assisted methods are generally successful in detecting arrhythmia extent from ECG readings early. The buzzword in computer-assisted clinical settings is branch of artificial intelligence. Computer-assisted arrhythmia forecasting approaches, particularly, are widely used machine learning methodologies. Most recent research is focused on the utilization of high-dimensional learning data sets to build machine learning models. The large dimensions of data points used for the machine learning techniques, on the other hand, frequently leads to false alarms. Though the few contemporary models endeavored to handle this by using multiple classifiers as ensemble model, they evince improved decision accuracy when trained on high volume of data. They do, however, frequently exhibit significant false alerting, with the training data representing the high dimensional data points of the enormous amount of training data provided. This paper discussed an ensemble learning approach that selects optimal subset of data-points by fusing diversity evaluation method