Riassunto analitico
It is an even more important task today as banks have been experiencing serious challenges and competition during the past decade with bank loan defaults. It concerns banks to limit potential default risks, screening the customer’s behaviour, financial history and financial background. Banks should control credit management thoroughly. Approving of loan requires the use of huge data and considerable processing time. Before granting loans, banks had to take numerous precautions such as performance of the firm by analyzing preceding year’s financial statements and history of the customer behaviour. The decisions of approving loans may become wrong and resulted major losses. It is essential for a bank to estimate the bank loan defaults. Estimation of this sort of defaults has been done by statistical techniques through decades and with respect to recent development within the field of Machine learning algorithms and business intelligence. Data mining provides many techniques to analyze huge volume of data and detect hidden patterns to convert raw data into valuable information.The research analyses is to find the accuracy of prediction on the bank loan defaults over the ten million customers with bank that which method from a selected set of Data mining techniques shows the best performance in default prediction with regards to selected model evaluation parameters The investigated techniques were Logistic Regression, Random Forests, Decision Tree, Neural Network and k-mean clustering. Predictive analysis is performed on bank loan dataset, with various data mining techniques, using SPSS Software. As such data analytics can provide solutions to tackle the current phenomenon and management credit risks. The results indicate that depending on the models generated, the significant and explanatory variables of the dependent variable bank loan status are different. Today the formation of accurate models has proved to be an excellent tool to support bank management decisions, to identify defaults and strategies towards consumer behaviour.
|