Credit is a belief that one is given to a person or other entity which is concerned in the future that will fulfill all the obligations previously agreed. The objective of the research is to do credit analysis to determine the feasibility of a credit crunch, through credit analysis results, to investigate whether the customer is feasible or not. The method used to predict credit worthiness is by using two models, i.e. C4.5 classification algorithm model and Particle Swarm Optimization (PSO)-based C4.5 classification algorithm model. After testing with these two models, it is found that the C4.5 classification algorithm generates a value of 90.99% accuracy and AUC value of 0.911 to the level diagnostics Classification Excellent, but after the optimization with C4.5 classification algorithm based on Particle Swarm Optimization accuracy values amounted to 91.18% and the AUC value of 0.913 to the level of diagnosis Excellent Classification. These both methods have different accuracy level of 0.18%.

Credit is a belief that one is given to a person or other entity which is concerned in the future that will fulfill all the obligations previously agreed. The objective of the research is to do credit analysis to determine the feasibility of a credit crunch, through credit analysis results, to investigate whether the customer is feasible or not. The method used to predict credit worthiness is by using two models, i.e. C4.5 classification algorithm model and Particle Swarm Optimization (PSO)-based C4.5 classification algorithm model. After testing with these two models, it is found that the C4.5 classification algorithm generates a value of 90.99% accuracy and AUC value of 0.911 to the level diagnostics Classification Excellent, but after the optimization with C4.5 classification algorithm based on Particle Swarm Optimization accuracy values amounted to 91.18% and the AUC value of 0.913 to the level of diagnosis Excellent Classification. These both methods have different accuracy level of 0.18%.