Classification of The Insured On The Basis Of The Customers Loyalty Through Using Data Mining Techniques

Abstract

Insurance companies ask for too much information about their customers and the information was saved in their data base for many years, customers and prediction of their requirements as well as estimation of future customers were all classified on the basis of experimental inferences and guessing of the insurance companies. Therefore, most of the insurance companies carried out new and intelligent researches and investigations about their own customers through using information technology tools. The main purpose of this research is to identify the loyal customers who have bought fire insurance policy and have extended it. The data available in information bank system of fire insurance in a private company has been investigated in the country during four years. The obtained results involve classification and grouping of fire insurances on the basis of decision tree by using Rapidminer software. After that, the test data are separated from train data randomly and it is determined that 70% of data are used for training and the remaining 30% of data are expended for testing.