A continuity is a kind of inter-transaction association which describes the relationships among different transactions. Since it breaks the boundaries of transactions, the number of potential itemsets and the number of rules will increase drastically. In this paper we consider the problem of discovering frequent compressed continuity patterns, which have the same power as mining the complete set of frequent continuity patterns. We devised a three-phase algorithm, COCOA, for frequent compressed continuity mining.
|主出版物標題||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|編輯||Jean-Francois Boulicaut, Floriana Esposito, Fosca Giannotti, Dino Pedreschi|
|出版狀態||已出版 - 2004|
|名字||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|