Study On Association Rules Mining Based On Time Stamp

With the rapid development of information technology, the amount of data needed to be processed is also growing. How to extract valuable information from a large number of data is a problem to be solved urgently in many industries. Data mining is from the people do not know in advance, random data to extract implicit in which, fuzzy, large, noisy, incomplete, but has the potential value of information and knowledge, the subject has formed a system, began to play a huge role in the daily life.Association rules is an important research direction in data mining. In this paper, the data mining technology and association rules mining are analyzed and studied, and a method of association rule mining based on time stamp matrix is proposed.The work of this paper is mainly in the following aspects:(1) A brief overview of the data mining technology, including the definition of data mining, classification, common technology and data mining steps, then the association rules definition, classification, mining method, mining procedures and other related theories are introduced in detail.(2) Detailed analysis of the advantages and disadvantages of the Apriori algorithm, the Apriori algorithm do the exhaustive elaboration, and through the concrete example detailing the Apriori algorithm an improved algorithm, and proves the superiority of the improved algorithm through experiments.(3) Describes the basic theory and algorithm of mining association rules based on timestamp and improved the support method of calculation and improve the target item set support values, and by combining the timestamp and the matrix and proposes a new algorithm for mining association rules based on time stamp of the matrix. The innovation of the improved algorithm is characterized by combine the timestamp and the matrix, and using the weight and to reduce the size of the transaction matrix, also solve the traditional mining some interest in the itemset support degree is low. Through the example analysis, based on timestamp matrix of association rules mining algorithm(BTMA) to mining based on matrix of the Apriori algorithm to mining frequent itemsets. Simulation experiment results show that the algorithm efficiency is superior to the one based on Apriori algorithm of matrix.