Analysis Comparison of The Classification Data Mining Method to Predictthe Decisions of Potential Customer Insurance

Current insurance business increasingly developed due to the increasing public awareness to insure and provide protection against various aspects of his life. Insurance service companies whose customers come from bank customers are very concerned about the quality of service to customers. Constraints the companies are difficulties in determining potential customers. Buy or No if the company could identify levels-levels to determine the potential value of the customer so the customer data can be classified. As for some of the criteria that are considered important in determining the potential client that is based on area, age and job. For it to be developed an application of data mining to determine criteria for the customer. Data mining techniques applied is Classification while the method used is the classification Decision Tree (decision tree) and support vector machine (SVM). There are three parameters of the test were used as evaluation system i.e. accuracy, precision and recall. From the results of the comparisons, the decision tree has a higher percentage than the SVM i.e. 86.37% accuracy, precision and recall 86.25% 90.14%. ROC curve and decision tree classification of diagnosis rate is excellent whereas the SVM good classification.