|
|||||
|
|
| 基于投影数据库的序列模式挖掘增量式更新算法 | |
| 引用本文： | 陆介平,刘月波,倪巍伟,陈耿,孙志挥.基于投影数据库的序列模式挖掘增量式更新算法[J].东南大学学报(自然科学版),2006,36(3):457-462. |
| 作者姓名： | 陆介平 刘月波 倪巍伟 陈耿 孙志挥 |
| 作者单位： | 1. 东南大学计算机科学与工程学院,南京,210096 2. 上海工程技术大学科研处,上海,200366 3. 南京审计学院审计信息工程重点实验室,南京,210029 |
| 基金项目： | 高等学校博士学科点专项科研项目，江苏省自然科学基金，国家审计局审计科研所资助项目 |
| 摘 要： | 针对序列模式挖掘中的增量挖掘问题,提出一种序列模式更新算法ISPBP.算法引入序列数据库结构来存储从原始数据库中挖掘出的所有项、最大频繁模式以及它们的支持数,采用间接拼接方法,只需处理增量数据库,避免了对更新后数据库的重新计算.对于因增量数据库新产生的频繁模式,利用了在增量数据库中出现的频繁项集来减小投影数据库,进一步提高了算法的效率.理论分析和实验表明,算法是有效可行的,并且增量数据库越大,算法在效率上的优越性越明显,算法ISPBP优于传统增量式更新算法. |
| 关 键 词： | 序列模式 数据挖掘 投影数据库 增量式更新 |
| 文章编号： | 1001-0505（2006）03-0457-06 |
| 收稿时间： | 10 10 2005 12:00AM |
| 修稿时间： | 2005-10-10 |
Incremental updating algorithm for sequence patterns mining based on projected database |
|
| Lu Jieping,Liu Yuebo,Ni Weiwei,Chen Geng,Sun Zhihui.Incremental updating algorithm for sequence patterns mining based on projected database[J].Journal of Southeast University(Natural Science Edition),2006,36(3):457-462. | |
| Authors: | Lu Jieping Liu Yuebo Ni Weiwei Chen Geng Sun Zhihui |
| Institution: | 1. School of Computer Science and Engineering, Southeast University, Nanjing 210096, China; 2. Scientific Research Office, Shanghai University of Engineering Science, Shanghai 200336, China;3. Key Laboratory of Audit Information Engineering, Nanjing Audit University, Nanjing 210029, China |
| Abstract: | Considering the problem of incremental sequence pattern mining,an incremental sequential patterns mining based on projected database(ISPBP) algorithm is proposed.Sequential patterns base is applied to the algorithm,which stores all items,maximum frequent patterns and corresponding support counts in original database. Instead of remining impertinently,ISPBP updates the frequent(items) and patterns found previously by implicit merging and discovers new patterns by projection database.Furthermore ISPBP decreases the projection database using the frequent items in the increment database.Theoretical analysis and experiments testify that ISPBP is efficient and effective.The larger the scale of database,the more prominent the algorithm's efficiency.ISPBP outperforms the conventional incremental updating algorithms. |
| Keywords: | sequence patterns data mining projection database incremental updating |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |