Exploiting Pattern Recognition using Chimp Optimization Algorithm with Machine Learning for Financial Crisis Prediction

Financial crises are typically the result of a combination of economic, political, and market factors, and they can be triggered by events that are challenging to foresee. Financial Crisis Prediction (FCP) has played a most important part in the financial industry. While predicting financial crises is challenging, some attempts have been made by researchers and organizations to use Machine Learning (ML) techniques to find potential risks and patterns that could be related to impending financial crises. ML techniques can analyze enormous quantities of data and recognize complex relationships that mayn't be instantly evident to human analysts. Therefore, this study offers the design of Enhancing Financial Crisis Prediction using the Chimp Optimization Algorithm with Machine Learning (EFCP-COAML) algorithm. The primary objective of the EFCP-COAML technique is to predict the automated and accurate prediction of the financial crisis. For accomplishing this, the EFCP-COAML technique comprises two major processes such as classification and hyperparameter tuning. At first, the EFCP-COAML method uses a Kernel Extreme Learning Machine (KELM) based prediction procedure. In the next phase, the COA is employed for the parameter selection of the KELM model which in turn enhances the predictive performance. An extensive set of tests were accomplished to demonstrate the superior accomplishment of the EFCP-COAML technique. The simulation results inferred the advancement of the EFCP-COAML approach over other present techniques with greatest accuracy of 94.01%.