Prediction of heart disease by hybrid ensemble deep learning tunicate swarm algorithm and improved elman neural network with feature selection

Abstract

The world is surrounded with lots of data collected from various newlinesources. Every human being and non-human things are now generating data, newlinethese are stored in volumes of storage devices. Many well-known newlineorganizations and academicians have done lots of research in the field of data newlinescience to analyze the huge volume of the data to generate non-trivial newlineinformation. Many new technologies have spur in the market to analyze the newlinedata. The dataset is collected from standard newlineUCl repository. Experimental results concluded that the integration of Linear newlineModel with RFM makes the simple estimation procedure with improved overall newlineaccuracy than the respective models. The proposed method compares the newlineprediction performance of few existing approaches in terms of parameters, newlinenamely, precision, recall and F1-score. HRFMILM improves the accuracy as newline87% than individual classifier. newlinexv newlineAfter this data is getting ready for clustering based on Density Based newlineSpatial Clustering of Application with noise (Clustering is done with the help of newlinehigh-level density points, core, and low-level density points). And once newlineclustering is being done then classification is done with Ensemble Deep newlineLearning and Tunicate Swarm algorithm. So Hybrid EDL-TSA algorithm gives newlineaccuracy with TSA are UCI 97.5% and CVD 98.33% respectively. By using newlineHybrid EDL-TSA gives effective prediction of heart diseases newline