A Study On Customer Churn Analysis In Securities Company Based On Data Mining

In recent years,the rapid development of information technology and the wide use of mobile Internet have profoundly affected and changed our lives,which is a big shock to traditional industries.For instance,even the financial industry has to face the impact and embrace opportunity both brought by Internet Finance.Internet Finance is invading and encroaching the domain once occupied by traditional finance.In the world of big data,in order to face the pressure of the Internet Finance,the traditional financial industry is in urgent need of restructuring and reforming.As a result,bank,security companies and insurance companies begin to take data assets very seriously.How to apply data mining concepts and methods in the financial industry has become an important research topic.This thesis is based on the data mining technology and internship experience in a certain security company.It concludes the application of data mining in security companies and focus on the customer churn analysis.This thesis consists of five sections.Section 1 summarizes the concepts and the influence of the big data era.Section 2 analyzes the development situations and trends of the Internet Finance.Also,it illustrates the link between traditional financial industry and Internet Finance.Section 3 concludes classic method of data mining and compares different common data mining softwares.Based upon real-world customer data from CRM system in one certain company,section 4 demonstrates the complete data mining process about customer churn analysis.Different methods have been used such as cluster,decision-making tree and Logistic regression.In the end,this application gives out a practical way to predict probability of customer churn in security companies.According to the predicted results,decision-making tree shows the best prediction accuracy,which beyond 80 percent.The last part in this thesis makes a summary and addresses some ideas about technology and application prospects of customer loyalty research in securities industry based on data mining.