Customer Credit Risk Assessment using Artificial Neural Networks

Abstract

Abstract-Since the granting of banking facilities in recent years has faced problems such as customer credit risk and affects the profitability directly, customer credit risk assessment has become imperative for banks and it is used to distinguish good applicants from those who will probably default on repayments. In credit risk assessment, a score is assigned to each customer then by comparing it with the cut-off point score which distinguishes two classes of the applicants, customers are classified into two credit statuses either a good or bad applicant. Regarding good performance and their ability of classification, generalization and learning patterns, Multilayer Perceptron Neural Network model trained using various Back-Propagation (BP) algorithms considered in designing an evaluation model in this study. The BP algorithms, Levenberg-Marquardt (LM), Gradient descent, Conjugate gradient, Resilient, BFGS Quasinewton, and One-step secant were utilized. Each of these six networks runs and trains for different numbers of neurons within their hidden layer. Mean squared error (MSE) is used as a criterion to specify optimum number of neurons in the hidden layer. The results showed that LM algorithm converges faster to the network and achieves better performance than the other algorithms. At last, by comparing classification performance of neural network with a number of classification algorithms such as Logistic Regression and Decision Tree, the neural network model outperformed the others in customer credit risk assessment. In credit models, because the cost that Type II error rate imposes to the model is too high, therefore, Receiver Operating Characteristic curve is used to find appropriate cut-off point for a model that in addition to high Accuracy, has lower Type II error rate.

Statistical Features Based Approach (SFBA) for Hourly Energy Consumption Prediction Using Neural Network

“…The graph shows almost similar patterns as observed with the data of 7 days. If we consider the histogram of the same, it has shown improvements in the prediction as compared to the data of 4 days, and 7 days, the increase in the number of data samples directly affecting the performance of the MLP [28][29]. The absolute error in the prediction of energy consumption during 14 days of a month has been represented in figure 8.…”

Analytical Assessment of Security Level of Distributed and Scalable Computer Systems

“…The malfunctions of JM, particularly the time delays during the analysis of critical situations and decisions making, especially in the case when the DCS process the confidential in formation can disrupt or even a completely stop the DCS functioning [31], [32]. Thus, the parameters analysis and the reducing of security risk for Job Manager operations is a very important task.…”

“…Taken countermeasures allo w reduce the number of potential vulnerab ilit ies in the computer system, and for this purpose, in accordance with the current safety policy in DCS, are imp lemented the protection mechanisms. However, even if the countermeasures are already implemented there may remain so-called residual vulnerabilit ies that form the residual security risk, wh ich can be reduced by the additional protective mechanisms [31].…”

“…In the case, when the rate of the security threats realizat ion is exceeded the reaction rate, the attack is launched on the DCS resources [33]. The analysis of the existing security threats allow determine the risks of safety violations for the co mputer systems resources and determine the countermeasures to neutralize the potential threats, by reducing the security risks to an acceptable level [31]. Taken countermeasures allo w reduce the number of potential vulnerab ilit ies in the computer system, and for this purpose, in accordance with the current safety policy in DCS, are imp lemented the protection mechanisms.…”

“…In general, the follo wing factors have impact on the security risks: the nu mber of nodes and link channels of DCS, the intensity of the security threats to DCS resources realization, the intensity of security mechanis ms activation, i.e., the rate of reaction to the hazardous events [1], [17]- [19], [31], [32]. In the case, when the rate of the security threats realizat ion is exceeded the reaction rate, the attack is launched on the DCS resources [33].…”