Bankruptcy Prediction using Emperor Penguin Optimizer with Deep Learning Model on Qualitative Dataset

Bankruptcy prediction and credit risk study are amongstthe most important problems in the domain of financial and accounting decision making. Evolving an effectual Classification Rule Induction (CRD) structure for bankruptcy prediction and credit risk examination at suitable time is vital to avoid the business communities being bankrupt. Classical statistical approaches and Artificial Intelligence (AD) play an important role in bankruptcy prediction and credit risks. Most previous research works can be executed on quantitative approaches but some studies are presented on qualitative approaches for improvising the efficiency of bankruptcy prediction techniques. This study focuses on the development of Bankruptcy Prediction using Emperor Penguin optimizer with Deep Learning (BP-EPODL) model on Qualitative Dataset. The projected BP-EPODL technique identifies the existence of bankruptcy or not. Here, the BP-EPODL technique exploits EPO algorithm for the choice of optimum features. For bankruptcy prediction, the BP-EPODL technique applies Long Short Term Memory (LS TM) model. At last, Adam optimizer is applied as a hyperparameter tuning technique for the IS TM model. The experimental validation of the BP-EPODL technique is tested on benchmark qualitative dataset. The obtained values inferred that the BP-EPODLtechnique has shown effectual results over other models.