Logistic Financial Risk Evaluation Based on Sparse Factor Analysis with Horse Herd Optimization and TensorFlow Decision Forest

In recent days, the logistics industry played an important role in global trade by facilitating the movement of goods and services across international borders. This logistic financial risks arise from uncertainties in the transportation system and supply chain disruptions. Traditional approaches for evaluating logistic financial risk had faced several challenges which includes regulatory compliance and supply chain disruptions. Therefore, this paper proposes Space Factor Analysis-Horse Herd Optimization-TensorFlow Decision Forest (SFA-HHO-TFDT) approach for logistic financial risk evaluation which is employed on financial risk dataset which taken from the China Stock Market & Accounting Research (CSMAR) database. This data is preprocessed by using min-max normalization to normalize the data since it is used for scaling the data into particular ranges. Then, SFA is employed for analyzing factor analysis to extract factors. After that, HHO is utilized for selecting optimal factors to enhance the performance of evaluation model accurately. Finally, TFDT is employed to evaluate optimal factors efficiently and logistic financial risks. This SFA-HHO-TFDT effectively reduced the supply chain disruptions and effectively achieved in terms of MSE (0.0029), MAPE (2.1987) and MAE (0.0489) respectively.