Dissertation > Excellent graduate degree dissertation topics show

Study on Data Mining Application of the Insurance Industry

Author: MiShuaiJun
Tutor: XiQin
School: East China Jiaotong University
Course: Statistics
Keywords: Data Mining Decision Tree Neural Networks Data flow Insurance Fraud Cross-selling
CLC: TP311.13
Type: Master's thesis
Year: 2010
Downloads: 450
Quote: 1
Read: Download Dissertation

Abstract


China's insurance industry has experienced rapid development in support of national policy, the insurance density increased to 736.74 yuan in 2008 from $ 0.46 in 1980 / person / people. Despite the rapid development, but the road is very flat and also incompatible with the level of economic development in China. Reform and restructuring, resulting in the domestic insurance market environment, a fundamental change, change the pattern of individual insurance companies to monopolize the market, led to a number of large insurance companies, involved in a number of small insurance companies, new insurance companies continue Join the new pattern. Coupled with the profitability and management capacity of the domestic insurance industry is far behind the developed countries, and the acceleration of the process of opening up more to increase domestic insurance business and the pressure to survive. Therefore, fast and good development of the Chinese insurance industry, to support economic development and building a harmonious society, become issues of mutual interest of the government, industry and academia. Technical overview of the theory and application of foreign insurance industry data mining research and analysis, combined with the operation and management of the domestic insurance industry and the level of information, this paper presents the application of data mining technology in the domestic insurance industry. Theoretical research and empirical research and analysis, we come to the feasibility and necessity of data mining technology has applications in the domestic insurance industry. Research and experimental conclusions reflected in the following two aspects: (a) the application areas of data mining technology in the insurance industry. (1) Customer Relationship Management. The application of data mining technology in customer relationship passenger and focus on customer service analysis and customer credit assessment. Customer service mining behavior analysis through to customer inquiries and complaints, according to the recommendations of the advisory information and complaints to identify the basic characteristics of such customers to carry out targeted customer service. Customer credit assessment analysis, data mining and customer credit rating information extracted with the corresponding customer characteristics, according to the characteristics of different credit rating have to provide appropriate services or monitor. Customer relationship management is a fairly generic concept CRM target only customer service and customer credit rating. (2) identification of target customers. Target customers identified three methods: questionnaire, statistical analysis, data mining method. By comparing the pros and cons of the three methods, pointed out the advantages of data mining technology to identify target customers in the massive data sets and high-dimensional space. (3) the policy cross-selling. Customers purchase a variety of insurance products in combination can be extracted through data mining features and behavioral characteristics, and provide guidance for insurance marketing. (4) customer retention and churn analysis. Using data mining techniques to extract the characteristics of the loss of customers, churn model, and apply the model to predict the probability of loss of existing customers, in order to improve service or adjustment of product structure. (5) Insurance customer fraud analysis. This article refers to the insurance fraud the unilaterally fraud, misconduct insurance customers insurance enterprises. Compared to the statistical analysis, data mining technology has obvious advantages in dealing with a small probability event. Processing of massive high-dimensional data through data mining method to extract the basic characteristics of fraudulent customers, establish a fraud warning model, real-time monitoring and assessment of fraud risk insurance customers. (B) the empirical analysis of data mining technology in the insurance industry. The basic features and functions of (1) data mining technology, as well as mainstream data mining algorithms. (2) the system architecture of data mining, SAS EM4.3 data mining capabilities, as well as to select the basic principles of data mining tools. (3) application of data mining technology in the insurance industry generally processes the data mining standard CRISP-DM. (4) empirical research and analysis for policy purchase predict \Include: data preparation process, indicators screening methods, data mining method selection, the model efficiency evaluation, and model deployment required precautions. Elaborated experiments algorithms: decision tree algorithm, neural network algorithm, and logistic regression analysis the basic principle. Compare the statistical methods and data mining methods. Algorithms applicability, advantages and defects, as well as the requirements of the data. (5) Based on the data of the SAS EM4.3 and MatLab2007 mining practice. The details of data mining tools from the application level, the application process, including explanation of key program code. Analysis of the data mining models how to embed other information management systems to enable them to self-learning and update, enhance predictive analysis capabilities.

Related Dissertations

  1. High Speed Frequency Measurment and Non-Linearity Correction of Frequency Modulated Capacitive Displacement Sensor,TH822
  2. Fault Diagnosis Method Based on Support Vector Machine,TP18
  3. A Study on Healthcare Product Marketing Based on Data Mining Technology,F426.72
  4. Gao Zhong-ying academic thought and experience and use of Bufei Decoction treatment of common diseases of the respiratory system drug law,R249.2
  5. Bing- thick academic thought and clinical experience and empirical studies apply to turtle soups treatment of chronic kidney disease,R249.2
  6. Research on Anti-periodic Solutions of Delayed Cellular Neural Networks without Assuming Global Lipschitz Conditions,TP183
  7. Designs and Applications of Fuzzy Synthetic Evaluation Models Based on Parallel Algorithms,TP18
  8. The Design and Implementation of Bicluster Data Analyzing Software,TP311.52
  9. Research on Clustering Algorithm Based on Mutation Particle Swarm Optimization,TP18
  10. Research on Fuzzy C-Mean Clustering Algorithm Based on Particle Swarm Optimization and Shuffled Frog Leaping Algorithm,TP18
  11. Research on Clustering Algorithm Based on Genetic Algorithm and Rough Set Theory,TP18
  12. Design and Implementation of Cross-selling System Based on Knowledge Reasoning,TP18
  13. Spectrum Variation of Vegetation in Yanzhou Coal Mine Area and Heavy Metal Stress Characteristic,X173
  14. Based on data mining research tax audit case selection,F812.42
  15. Community-oriented education, personalized learning system and its implementation,TP391.6
  16. Association rule mining based Intrusion Detection System Research and Implementation,TP393.08
  17. Data warehouse technology in the banking customer management systems research and implementation,TP315
  18. Design to E-learning System in Senior Vocational School Base on Moodle,TP311.52
  19. Design and Development of Teaching Quality Assessment System Based on Data Mining,TP311.13
  20. The Application of Association Rules Algorithm in Higher Vocational Colleges’ Endorsement of Impoverished Students,G717
  21. Based on Data Mining Technologies in Urban Water Supply Analysis and Decision,F299.24;F224

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer software > Program design,software engineering > Programming > Database theory and systems
© 2012 www.DissertationTopic.Net  Mobile