Hybridized Gradient Lstm And Rnn For Ai-Iot Assisted Financial Disaster Prediction In Fintech Environment

Abstract

FinTech is a separate terminology that analyzes the financial technology industries inside a wider series of commands for firms using IT tools. As the Internet of Things (IoT) grows exponentially, AI-enabled flexible IoT is the future of financing. IoT's depth has likely changed the financial industry today, and it could quickly become a dominant instrument in the future. The use of AI and IoT can greatly enhance financial data extraction and customer support. Financial disaster prediction (FDP) seems to be a process of assessing a corporation's economic position. The Fintech index's forecast could also allow shareholders to prevent costs and help financial managements. So, we propose a novel hybridized gradient long short-term memory and recurrent neural network (HG-LSTM-RNN) in a FinTech setting. The financial information of global businesses is acquired in the beginning phase using IoT applications like mobile phones and computers. To select the optimum features, we utilize the flexible chaotic henry gas solubility optimization (FCHGSO) approach. Furthermore, the proposed approach is used to classify the gathered financial data with the greatest prediction rate. The performances of the proposed approach like sensitivity, F-score, accuracy are examined and compared with existing approaches to prove our research with the greatest effectiveness. The findings of those performances are depicted in graphical representation using the MATLAB tool.