Approach To Data Mining Based On Rough Neural Network And Its Application

In the recent years, with the quick development of database technique and the extensive application of database management system, enterprise have accumulated more and more data in their database or data warehouse. However, there are many important information is concealed in the numerous data. In order to use these data more efficiently and provide accurate decision-making project for the marketing and development of the enterprise, the enterprise' s leaders hope to analyze the data with higher level. But people discriminate and extract the valid and strong characteristic from numerous data is still a problem without complete solution currently.Firstly, this paper introduces the fundamental of Data mining and its application in marketing, and summarizes the application of Rough sets and Neural network which are two important methods in Data mining. Secondly, it introduces the classical Pawlak Rough sets model, which is a uncertain and vague conception based on equivalence relation and expresses by upper and lower set approximations. Owing to the strong condition of equivalence relation, there exist some limits, and It has been proved to be NP-Hard to find all reductions and a minimal reduction. As the problem of attribute is one of core problems in Rough sets theory, this paper puts forward an improved algorithm based on GENRED reduction algorithm, what' s more, the improved algorithm is applied into client purchase analysis field. In what follows, it introduces the variable precision Rough sets model and its application in enterprise management actuality analysis, and the probability Rough sets model and its application in production sort, and the imperfection information system Rough sets model and its applicationin production sale respectively. For the enormous data and noise disturbance when mine data from large database or data warehouse, only using the Rough sets in data mining would decrease the classification precision, and only using the Neural network in Data mining would make the structure of Neural network complex and the training overtime, Finally, This paper introduces the flow of data mining method and the marketing structure based on Rough Neural network. Then based on this, the combination of Rough sets and Neural network is applied into marketing product sale field.