Design of efficient feature selection and ensemble learning approaches for covid 19 forecasting

Abstract

It is widely known that a quick disclosure of the COVID-19 can help to reduce its spread dramatically. Transcriptase polymerase chain reaction could be a more useful, rapid, and trustworthy technique for the evaluation and classification of the COVID-19 disease. Currently, a computerized method for classifying data can be crucial for speeding up the detection while the COVID-19 epidemic is rapidly spreading. In the earlier work, different techniques like auto encoder, Linear Regression (LR), Exponential Smoothing (ES), Long Short-Term Memory (LSTM) and Support Vector Machine (SVM) have been used for the prediction of COVID-19 future. All of them provides poor classifications results due to irrelevant classification methods. Also, data used in the classification consists of several features while modeling development it reduces Attention deficit hyperactivity disorder classification performance. To solve the above-mentioned problem in first phase introduced a Linear Decreasing Inertia Weight based Cat Swarm Optimization with Half Binomial Distribution based Convolutional Neural Network (LDIWCSO-HBDCNN) approach for COVID-19 forecasting. In this proposed research work, the COVID-19 forecasting dataset is taken as an input. The input data normalization is done by using min-max normalization approach. Optimal features are selected by using Linear Decreasing Inertia Weight based Cat Swarm Optimization (LDIWCSO) algorithm to improve the classification accuracy. In LDIWCSO algorithm, inertia weight factor is added to improve the convergence of Cat Swarm Optimization (CSO). newline newline newline