Title :
An Algorithm of Dynamic Associate Rule Based on Sliding Windows
Author :
Meng, Zhiqing ; Shao, Baosheng ; Jiang, Min
Author_Institution :
Coll. of Bus. Adm., Zhejiang Univ. of Technol., Hongzhou, China
Abstract :
It is well-known that the traditional association rules with time can´t be found out in many algorithms. First, according to concept of dynamic association rules, this paper analyzes the disadvantages of traditional association rules. Then, the concepts of dynamic association rules based on sliding windows and the definition of time vector representation of dynamic association rules are put forward. So, the dynamic association rules can measure out how association rules change over time. Next, a kind of algorithm to mine dynamic association rules on sliding window is proposed. Finally, numerical experiments show that the proposed algorithm is efficient for finding out dynamic association rules with time.
Keywords :
data mining; dynamic association rule; sliding window; time vector representation; algorithm; dynamic association rule; sliding windows; time vector;
Conference_Titel :
Computational Intelligence and Security (CIS), 2010 International Conference on
Conference_Location :
Nanning
Print_ISBN :
978-1-4244-9114-8
Electronic_ISBN :
978-0-7695-4297-3
DOI :
10.1109/CIS.2010.18