DocumentCode :
2889525
Title :
Research on Dynamic Generating Algorithms of Large Itemsets of Distributive Data Mining Architecture
Author :
Fang, Ying-Wu ; Wang, Yi ; Li, Peng-yang ; Lu, Yan-jun ; Zhao, Xiu-bin ; Xu, Hui
Author_Institution :
Telecommun. Eng. Inst., Air Force Eng. Univ., Xi´´an
fYear :
2006
fDate :
13-16 Aug. 2006
Firstpage :
1314
Lastpage :
1319
Abstract :
Dynamic generating algorithms of association rules mined large itemsets are presented in this paper. According to the distributive data mining calculation architecture, database is replaced by an order set enumerate tree, and the information of all transactions are kept in the dynamic generating trees. Meantime, the generating enumerate trees flowing a local node of transaction orderly is ensured. The storage and communication traffic are reduced greatly. Therefore, the local space is saved, and the disk operation is reduced through the generating algorithms of large itemsets. By examples and performance analysis of the dynamic generating algorithms introduced, the store space of processed nodes is cut down, and the calculation time of support is also reduced through the traversal process of tree. Thereby, the calculation efficiency of search is improved greatly
Keywords :
data mining; distributed databases; tree data structures; very large databases; association rule; distributive data mining calculation architecture; dynamic generating algorithm; large itemset; order set enumerate tree; Association rules; Cybernetics; Data engineering; Data mining; Heuristic algorithms; Instruments; Itemsets; Machine learning; Machine learning algorithms; Machinery; Relational databases; Transaction databases; Data mining; association rules; generating algorithms; set enumerate trees;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Machine Learning and Cybernetics, 2006 International Conference on
Conference_Location :
Dalian, China
Print_ISBN :
1-4244-0061-9
Type :
conf
DOI :
10.1109/ICMLC.2006.258659
Filename :
4028267
Link To Document :
بازگشت