The Design And Realization Of Classification System Of One-to-One Marketing Based On Data Mining

With the fast expansion and deepening of the market, the constant development of enterprises, the abundance of the products, the customers' choice is expanded greatly. In order to satisfy the customers' individualized demand that increases day by day, enterprises must change the traditional marketing mode, which means change the marketing centre from regarding market share as of the focal point of marketing into regarding customer's share as the centre. One-to-One marketing is produced under such a demand, However, One-to-One marketing have real value, not just an idea, thanks to the flourishing development of Internet and IT new technology. Even so, the existing circumstances of one-to-one marketing is not optimistic because of lacking support of information technology. One-to-One marketing must implement based on data base technology and data mining technology. Understand customers' differences through Write down and classify the customers' information constantly. Then, grasp the customers' information dynamic and develops the activity of marketing meaningful. Among them, classifying customers' information is the most important problem for one-to-one marketing. The purpose of the thesis is to set up a classified system suitable for implement of enterprises.This thesis elaborate the generation and development of one-to-one marketing, introduce the concept and main points of one-to-one marketing, and describe the implementation step of one-to-one marketing, analysis the relationship between one-to-one marketing and traditional marketing. Through introduce the theory and the existmg circumstance of development for one-to-one marketing, it propose the existing problems of one-to-one marketing, and point the great meaning that using data mining technology to implement one-to-one marketing. The main content of the thesis is the study of classification system. The thesis introduces two key technology: data base technology and data mining technology. Then, it propose the differences between data mining and traditional analysis method, introduce some classification method of data mining technology, and introduce the theory of C4.5 algorithm specifically. Based on the analysis of the theory and combined the characteristics of one-to-one marketing, propose main aim and system function of classification system for one-to-one marketing, and give the details about the composition of classification system.The thesis detailed introduce the composition of classification system, including overall structure, data base design and specific implement of system functions. In the data base design, detailed introduce requirement analysis, concept design and logic design. In the implement of system functions, detailed introduce data purging, data filtering, the structure of classified table, the structure of classification model, pruning and the selection of classification rule, and using inspection principle to judge the correctness of classification model. Based on the analysis of theory and combined specific example describe the structure of classification model. Finally, the thesis introduce some tools of data mining, and analysis the relationship between classification system and other subsystem in the marketing system, including research subsystem and decision subsystem.The innovation of the thesis is represented by the structure of classification system and implement of functions, based on data base and data mining technology, and combined the study of one-to-one marketing.