Life insurance industry data mining application analysis life insurance is an important branch in the insurance industry, with huge market development space, so with the opening of the life insurance market, the intervention of foreign companies, the competition is gradually upgraded, and the group is competing. How to maintain its own core competitiveness, so that you have always been invincible, and it is a problem that each company must face. The application of information technology is undoubtedly one of the effective means of improving the competitiveness of enterprises. After years of development, the life insurance information system has gradually matured, and has accumulated a significant amount of data resources. It provides a solid foundation for data mining, and the knowledge is discovered through data, and is used for scientific decision-making increasing life insurance companies. Pay attention. Data Mining Data Mining (DM) refers to a process of extracting, useful information and knowledge from a large amount of incomplete, noise, blur, random data. Its expression is in the form of concepts, rules, pattern (PATTERNS). At present, there have been many mature data mining methodology in the industry, providing an ideal guidance model for practical applications. CRISP-DM (Cross-Industry Standard Process for Data Mining) is recognized, one of the more affected methodology. CRISP-DM emphasizes that DM is not only the organization or presentation of data, not only data analysis and statistical modeling, but a complete process from understanding business needs, seeking solutions to accepting practical tests. CRISP-DM divides the entire mining process into the following six stages: Business understanding, Data Understanding, Data Preparation, Model, Evaluation, and Reployment . Business understanding is to understand the background of business operations, business processes, and industry background; data understanding is an understanding of existing enterprise application systems; data preparation is to remove a sample data subset of samples related to the problem. Modeling is based on the understanding of business problems, on the basis of data preparation, select a more practical mining model to form a conclusion of mining. The assessment is the conclusion of the excavation in the actual inspection. If the expected effect is reached, the conclusion can be released. In actual projects, data understanding, data preparation, modeling, and evaluation are not one-way operation, but multiple times, and constantly revisit and revisit multiple times. Industry Data Mining After years of system operation, life insurance companies have accumulated considerable policy information, customer information, transaction information, financial information, etc., and there have also occurred a large scale database system. At the same time, the data is concentrated for the promotion of the original business level and the expansion of new business, which provides a rich soil for data mining. According to the CRISP-DM model, data mining should first be done to understand the business, find the goals and problems of data mining. These problems include: the agent's selection, fraud identification, and market segmentation, the market segmentation has high guiding significance to the business strategy of enterprises, it is related to whether the company can survive and develop, corporate marketing strategy formulation The primary problem of realization. In response to life insurance business, we can classify customer groups from different angles to form various customer distribution statistics, as a basis for administrators. Starting from life insurance products, analyzing customers' preferences for different dangers, guiding agents, is a mining idea that is easier to implement. Due to the different domestic economic development, the provinces differ, so they must be limited to analytical data in an economic level.
