Arithmetic Of Association Rules Mining Based On Dynamic Data

Data Minging distills useful message from a mass of data. It is a new research field involving several branchs. The association rule is one of the most important domains among them, which is used to find the interesting relations involved in items or attributes of database. These relations are undiscovered before and it cannot be gotten through logic operations or statistic methods of traditional database operation techniques. Therefore, mining association rule doesn't base on self-attributes but on co-appearance characteres among items of database.Rules are thought as invariable in the former mining arithmetic, that to say, the rules are just static rules. In fact, data characteristics and rules may take huge changes during the process of time, therefore maintaining the availability of association rules mining is essential especially. This paper is based on the theory of association rules mining and objected to the practical application of the industrial process. The effective strategy of association rules of dynamic data was discussed. Based on the FP-Growth arithmetic, The TFP arithmetic is presented and it can optimize the production technics to accelerate the higher production efficiency.Based on the background of industrial production and the definite view of characteristics classification of association rules mining, several techniques in different phases of data mining processes about frequent items of association rules mining of dynamic data are investigated in this dissertation. Additionally a practical application of data mining technique is evaluated and it is proved fine.