A Study On Data Mining-based Customer Intelligence

Along with the economical globalization and the high and new technology fast development, the world entered a brand-new knowledge economy time. The competition environment, which the enterprise locates, has had deep transformation. Some market starts to decline, the competitors become time of growth, and the product quality and the price assimilates day by day. Customer's consumer behavior becomes daily rationalized and personalized. Customer-driven becomes the main characteristic of the new market. In this case, traditional CRM is abolished for unable to satisfy the enterprise's need of keeping and using customer knowledge property. Customer Intelligence becomes a new strategy for the enterprise to maintain its core competitive power.Customer Intelligence(CI) is the strategic activity that the enterprise fully utilizes various intellectualized technologies, converts customer data into customer knowledge, and takes advantage of the knowledge to improve customer relationship, make customer decision and promote customer value. Introduce Customer Intelligence into the customer management will enable the enterprise to understand consumer patterns reasonably, be initiative in the customer relationship, and forecast the customer behavior and the market trend. Customer Intelligence is inosculated with the enterprise management, implanting the ideas of"take the customer as the center"into the organization, and enabling it to have a stronger feasibility. Customer Intelligence involves the entire process from producing to distributing and using the customer knowledge, emphasizes the knowledge fluidity and the validity in policy-making.Being a new concept having characteristic of disciplines overlapping, Customer Intelligence has absorbed the research results of customer relationship management, knowledge management, business intelligence, market marketing etc, and takes many high-level knowledge technology as the support, such as knowledge engineering, artificial intelligence, statistics, computer technology and so on. Data Mining is the most essential technology in the Customer Intelligence system. Data Mining(DM) is the process of extracting concealed, unknown and latent useful knowledge and pattern from massive incomplete, noisy, fuzzy and stochastic data. In the commercial domain, Data Mining may be helpful to analyze known facts, uncover and forecast unknown results. It can analyze the key business factors and discover the business tendency. Because customer knowledge is the main attention object of Customer Intelligence and Data Mining is the core of knowledge discovering and acquiring, the CI and the DM have close associativity. Only when Data Mining is reasonably integrated into the Customer Intelligence process, can realize the goal of prompt and effective customer knowledge extraction from multitudinous customer data.This paper has conducted the research from different angles on Customer Intelligence theory, method and application using kinds of Data Mining technologies. Its goal lies in the effective combination of Data Mining technology and customer business target, the construction of Data Mining-based Customer Intelligence theory and method system, to sustain the customer knowledge acquiring, innovation and the customer decision-making optimization.The main content in this paper includes six parts as follows:(1) Customer Intelligence Theory OutlinesThis part generally introduces the theory of Customer Intelligence. In the foundation of summarizing correlated research results, it has carried on a thorough analysis to the key issues in the elementary Customer Intelligence theory frame including background, definition, essence, theory base, technology base, functions and advantages.The paper uncovers the reason and background in which Customer Intelligence is developed from three aspects: the change of the market competition environment, the establishment of the new"knowledge = property"idea and the challenge and the opportunity brought by information technology development. The paper points out from a strategic level that Customer Intelligence is a strategic activity, which produces and uses customer knowledge by applying intelligent technologies. The paper analyzes the Customer Intelligence's connotation from two key aspects: the customer knowledge and the customer value, and points out the Customer Intelligence's essence, which is making decision and taking action by innovating customer knowledge, helping the enterprise to obtain and enhance the customer value, strengthening its competitive advantages. The paper thinks customer theory, knowledge management theory and business intelligence theory are the three main theory bases of Customer Intelligence. And Data Mining, data warehouse and OLAP, and artificial intelligence are its three support technologies. Customer Intelligence has great significance in enterprise. Its main functions include data management, information analysis, knowledge discovery and enterprise modeling. Through the discovery of customer knowledge, Customer Intelligence can provide scientific basis for customer business analysis and decision-making forecast.(2) Data Mining-based Customer Intelligence ArchitectureThis part analyzes each kind of essential factor in the realization of Customer Intelligence from different aspects, constructs the Data Mining-based Customer Intelligence architecture.The paper points out that Customer Intelligence is a multi-dimensional system, involving various strategic management factors and technical factors in the enterprise. The Data Mining-based Customer Intelligence architecture is composed of five layers. The theory foundation layer provides theory instructions for the Customer Intelligence activities. It contains principles and methods of customer analysis, customer knowledge management and customer relationship management and so on. The paper discusses with emphasis about the customer analysis theory supporting the customer knowledge mining, including customer life cycle analysis, customer value analysis, customer satisfaction analysis and customer loyalty analysis. The data storage layer is the physical foundation of the Customer Intelligence. It is the resource of customer data, including enterprise interior and exterior information platforms and database systems. The paper has analyzed the data constitutions and characteristics in different data sources. The information analysis and integration layer provides a comprehensive and unified customer information view for the Customer Intelligence. The paper has analyzed the primary coverage of the customer information view, and explored the way and the methods to realize customer information integration through data warehouse. The knowledge discovery layer is crucial to Customer Intelligence activities. It has the closest relations with Data Mining, carrying on knowledge analyzing, knowledge mining, knowledge classifying and knowledge modeling. The paper has discussed some typical Data Mining methods and algorithms including statistical regression, clustering, decision tree, neural network, association rules mining and genetic algorithm. The strategic management layer is the topmost layer in the Customer Intelligence architecture, which collects a set of ideas, strategies and methods of customer relationship management and enterprise decision. The paper has studied from two aspects of strategic decision and tactical decision on knowledge-guided customer management decision in enterprise.(3) Data Mining-based Customer Intelligence MethodologyThis part studies the entire realization process of the Data Mining-based Customer Intelligence. Based on the analysis and description of enterprise business problem, it studies the patterns, steps and methods of building and applying Data Mining models. It summarizes the ordinal steps of Data Mining practice, and constructs a set of effective and scientific methodology.They paper points out that the Data Mining-based Customer Intelligence flow is a virtuous cycle of knowledge activities. It is composed of four phases, that is, discover customer business opportunity, apply Data Mining to form customer knowledge, make decisions and take action, and appraise action result. At an engineering angle, the paper has discussed the key activities and technical factors in the three main processes of mining goal analysis, data preparation and data modeling. In the process of determining customer business goal and Data Mining task goal, some major activities should be carried on, including recognizing the people related to business, recognizing the business requirement, setting the business analysis goal, analyzing the Data Mining environment, setting the Data Mining goal, form the Customer Intelligence project plan. In the process of data preparation, the right customer data should be chosen and examined at first, the problems discovered should be repaired, the data should be transformed into customer information suitable for mining, and finally the data should be partitioned into modeling set. In the process of Data Mining modeling, it should firstly analyze the features of input/output variables and choose proper methods and algorithms to build initial model, and then test, ameliorate and optimize the model, finally evaluate the model application effects.(4) Data Mining-based Customer Knowledge Acquiring and Forecast Based on the theory and method research in the front, this part, taking the customer life cycle as the main line, applies Data Mining to realize customer knowledge analysis, acquireing and forecast in different phrase of customer relationship management.The paper has studied the customer groups and business activities which the enterprise faces in different relation stages, as well as the combination of Data Mining and various business activities. It has explored Data Mining implementation ways in five significant fields of customer profit ability analysis, customer response forecast, customer segment, customer increment expense forecast and customer churn forecast, especially discussed the crucial issues of data, modeling and result using.(5) Data Mining-Based Customer Intelligence Application InstanceThis part applies the theory research achievement to practice, and introduces an application instance of customer analysis using Data Mining in Customer Intelligence. This instance utilizes SAS Enterprise Miner, and realizes the common response forecast in market promotion. The paper has discussed and demonstrated in detail the entire process from preparing data, establishing model to applying model.(6) Summary and ProspectThis part summarizes the research of the paper and forecasts the trend of Customer Intelligence.Through the research of Data Mining-based Customer Intelligence in this paper, we can draw the following three conclusions:â‘ Customer Intelligence is one kind of strategic choice for the enterprise to ensure its unique competitive advantage.â‘¡Customer Intelligence manifest the synthesizing capability of enterprise customer knowledge management.â‘¢Customer Intelligence is an integrated application of high-level information technologies in commercial domain. The paper also points out that under the impetus of policy, market and technology, Customer Intelligence will welcome a climax of development. Its research emphasis will concentrate in the intellectualization of customer knowledge management, the effective combination of customer knowledge and enterprise strategic management etc.