Title :
Discovering Frequent Itemsets over P2P Network
Author :
Song, Wei ; Gao, Lei ; Wang, Junshan ; Li, Jinhong
Author_Institution :
Coll. of Inf. Eng., North China Univ. of Technol., Beijing, China
Abstract :
Mining frequent item sets over P2P network is a challenging problem. The obstacles mainly lie in numerous frequent item sets and huge communication cost. To solve the two problems, a maximal item set mining algorithm P2PMaxSet is proposed. Firstly, only maximal item set is considered, which reduces the number of item sets greatly. Secondly, mining results are only interchanged between neighbor nodes, which save communication cost. Finally, adjust strategies are discussed for dynamic environment. Experimental results show P2PMaxSet is not only accurate but also with lower communication cost.
Keywords :
data mining; peer-to-peer computing; P2P network; P2PMaxSet; adjust strategy; frequent item set mining; frequent itemset discovery; maximal item set mining algorithm; neighbor nodes; Approximation algorithms; Data mining; Distributed databases; Heuristic algorithms; Itemsets; Nickel; Peer to peer computing; P2P network; data mining; maximal frequent itemset;
Conference_Titel :
Internet Computing & Information Services (ICICIS), 2011 International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4577-1561-7
DOI :
10.1109/ICICIS.2011.104
