An approach for Credit-Scoring using Swarm intelligence for feature selection and PSO trained ANN based classification

As the financial system expanded, the credit scoring process changed in a way that has attracted more interest from scholars and businesses. Artificial intelligence technology based on predictive classification has changed how credit scores are calculated. These decision-making processes are taken with the help of various data mining algorithms to predict if the client is part of a suspicious group which is more likely to cause losses. In our proposed model (SIFS-PNN), we have used swarm intelligence-based algorithms to find relevant data along with a fine-tuned artificial neural network to successfully classify, whether or not, a client is a part of a sensitive group of credit lines. To do this, we will pick features using various swarm intelligence algorithms and fine-tune our ANN using Particle Swarm Optimization algorithm to perform credit classification. We also performed a comparative study with multiple previous researches to show how the suggested approach has outperformed previous results