Customer churn analysis with machine learning in insurance sector

In today's fast-growing and competitive world, the effort to acquire new customers and the effort of not losing the existing customer and its cost is more than the cost, there are thoughts to think companies to make an existing master. In the company where there are strong competitors, analytical research has been carried out on increasing the loyal customer portfolio that meets the consumers by choosing a company for a service or product that prefer a company for a certain service or product, and providing a product service preferred by the customer. Finding customers with the aim of researching campaigns and behaviors, customer loss analysis aiming to increase the satisfaction of these customers can be one of the most important stages of strategic decision making and planning. This phone is about the socio-demographic response of a customer such as the number of customers such as telecommunication, banking, online trade and the amount of income in the insurance industry, as well as the characteristics of the vehicle brand and model used, as well as the socio-demographic answer such as age, gender, place of birth. Desicion Tree Algorithm, Random Forest Algorithm and K-Nearest Neighbor (K Nearest Neighbor) Algorithms, attributes determined from machine learning algorithms are predicted by machine learning algorithms. With the Random Forest Algorithm, which gave the most successful results in the study, this study was included in a class and allowed to be continuously performed by the end user.