作者: Phillip Shi-lung Yu , Bob Chao-Chu Liang , Ming-Syan Chen
DOI:
关键词:
摘要: For mining causality rules in an event database, the are obtained by iteratively generating candidate and counting their occurrences database. Newly identified used to generate next set of be evaluated, increasing size consequential events triggered triggering and/or number events. The preferred embodiment uses iterative approach deriving order sizes sizes. detection occurrence a rule sequence is handled as sub-sequence matching problem using novel hierarchical method improve efficiency.