| An efficient algorithm to generate candidates in discovering frequent episodes |
| |
| 作者姓名： | 邓勇 Liu Qi Li Yixue |
| |
| 作者单位： | School of Electronics ＆ Information Technology, Shanghai Jiaotong Universily, Shanghai 200030, P.R.China |
| |
| 基金项目： | 国家高技术研究发展计划(863计划) |
| |
| 摘 要： | 0 IntroductionData mining is widely used in many research fieldssuch as decision supporting systems1], bio-informationanalysis2]and knowledge engineering3-5]. Most data col-lected from scientific experiments or telecommunicationnetworks have inherent sequential nature inthemand canthus be abstractly viewed as a sequence of events . Onebasic problemin miningsuchevent sequencesis discoveryof recurrent combinations of events , which are calledepisodes. Once frequent episodes are discovered,rul…
|
| 关 键 词： | 事件顺序 WINEPI 搜索空间 算法 候选人 |
| 收稿时间： | 2005-03-16 |
An efficient algorithm to generate candidates in discovering frequent episodes |
| |
| Deng Yong,Liu Qi,Li Yixue.An efficient algorithm to generate candidates in discovering frequent episodes[J].High Technology Letters,2006,12(1):109-112. |
| |
| Authors: | Deng Yong Liu Qi Li Yixue |
| |
| Abstract: | One of the important steps in mining event sequences is to find frequent episodes. Once the frequent episodes are discovered, rules about temporal relationships can be derived. In this paper, an efficient algorithm for discovering frequent episodes is presented based on the level-wise search algorithm WINEPI.The proposed algorithm gains better candidate generation quality by introducing a new Lemma to help to target the combinations of episodes that are interesting in the next level and thus reduces the execution time. Experimental results on artificial and real data show the enhanced efficiency of the algorithm. |
| |
| Keywords: | frequent episodes event sequence WINEPI new Lemma search space candidategeneration |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
