A new fuzzy clustering algorithm of User Transaction

A new fuzzy clustering algorithm of User Transaction named MTCM is presented based on the analysis of the drawbacks of traditional methods. Firstly, statistics of the page access frequency and time length is kept in user transaction and map them to the continuous access time attribute value. Then, dimensions of transaction is reduced. Finally, according to the calculation of fuzzy similarity between any two transactions, a fuzzy similarity matrix is got to achieve transaction clustering. This algorithm is proved to have better accuracy, fewer CPU time and better expansibility than others.