Research Of Data Mining On Telecomunication Sales

With the quick development of the database technology and with the wide application of the database system, more and more data have been piled up. The overfull data have hided much more important information, and people wish to analyze them in higher level to make good use of them. All of the above lead to a new technology named Data Mining emerging as the times require. Data Mining is deemed as an interdisciplinary study composed of database theory , statistics, machine studying theory and Nerve Net method. Although Data cMining is based on database theory, it is faced to application from the beginning. It is applied not only to the simply search in specific database, but also to analyze and reason the data from microcosmical to macroscopical. It is used to instruct people to answer the actual questions and to find the interactive relationship between the different affairs , even to forecast the future with known data. All these characteristics determine the system and the method of Data Mining.In this paper,the history and development of Data Mining are fully described,conceptions and characteristics theoretically discussed,and the significance and objectives explicitly determined.Then through analysis of Data Mining technology and the research of the judge standards of the Data Mining's tools, it is further explored that data model will be built with different Data Mining arithmetic according to the urgent questions in the telecom marketing, and the research objective,research methods and research context for the program are finally defined.One of the important research work in this paper is discussing the application that the decision tree is used in the telecommunication marketing field. It is building customer losing model using Data Mining method and decision tree arithmetic that mainly according to the character and the consuming behavior of the customer losing or not in the telecommunication enterprise. Marketing policies are also established according to customer losing model. In this portion , we discuss the construction theory of the decision tree and the shaving theory of the decision tree avoiding the excessive training. Then this paper introduces the application of clustering Methods in the telecommunication marketing field.Firstly, this paper introduces the theories and classification of the clustering arithmetic. Secondly, the paper introduces the experiments which mark off the telecommunication customer's level through the irregulation and external information with Nerve Net arithmetic, and carry out the marketing combination policies and different service system according to the classified results. Lastly, the paper discusses how to deal with the peculiar exemple in the clustering analysis.