Research And Application On The Intelligent Algorithm Based On Enterprise Data Mining

With the emergence and the popularization of Internet and the rapid pace of information technology, the mature of database technology, computer, network and communication mixed together, new generations of integrated information systems appears. Due to the openness, the interconnectedness and the sharing of the system itself, the enterprise has accumulated mass of data."Rich data, poor information". A pure data query and transaction cannot meet the needs of users, how to extract the valuable information and knowledge from the vast amounts of data. Thus, to discover the development trend of the things, and help enterprises to adjust competition strategy continuously and make enterprise invincible in competition, this is the main task of data mining.Smart client system is an important part of the enterprise information system based on the SaaS platform; it is also an interaction platform between the data mining system and the enterprise information system. First of all, to the research target of the client of enterprise system based on the SaaS platform and to the advantages and disadvantages of C/S mode and B/S mode, a new kind of intelligent client structure has been presented. the key techniques of smart client are introduced in detailed ,the core questions of data security and data offline of smart client are resolved; Secondly, we conducted in-depth understanding of the system structure, function module and data mining process of the weak platform. An improved clustering technology and improved RBF network are designed on the decision tree of machine learning, on the basis of, and the results of the experimental testing show the effectiveness and capableness of the improved algorithm.Tirdly, according to data mining of application of enterprise customer churn, taking enterprise customer information as the research target, a set of effective scheme is formed; customer information is analyzed by using intelligent algorithm. And the models are evaluated by the WEKA platform; the results show that it is feasibly and effective.Finally, we make a summary of the full text and explore new fields of development trend of data mining in enterprise.