<-上一篇/Previous Article 下一篇/Next Article->

[1]史忠植,尹超,叶世伟.一种支持时间序列数据的CBR检索算法[J].智能系统学报,2007,2(1):40-44.
　SHI Zhong-zhi,YIN Chao,YE Shi wei.A CBR algorithm supporting time series data[J].CAAI Transactions on Intelligent Systems,2007,2(1):40-44.

点击复制

一种支持时间序列数据的CBR检索算法

PDF下载 HTML

《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷: 2 期数: 2007年第1期页码: 40-44 栏目: 学术论文—人工智能基础出版日期: 2007-02-25

Title:: A CBR algorithm supporting time series data

文章编号:: 1673-4785(2007)01-0040-05

作者:: 史忠植¹，尹超^1，2，叶世伟²; 1.中国科学院计算技术研究所智能信息处理重点实验室，北京100080；
2.中国科学院研究生院信息科学与工程学院，北京100039

Author(s):: SHI Zhong-zhi¹, YIN Chao^1,2,YE Shiwei^2; 1.Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100080, China；
2. School of Information Science and Engineering Graduate University of Chinese Ac ademy of Sciences, Beijing 100039, China

关键词:: 基于范例的推理; 时间序列数据; 相似度比较

Keywords:: casebased reasoning; time series data; similarity c omparison

分类号:: TP399

文献标志码:: A

摘要:: 探讨了如何为CBR（基于范例的推理）增加对一种特殊的范例类型——时间序列数据的支持.分析了基于谱分析的时间序列相似度比较算法不适用于CBR检索的缺点，并在此基础上设计了一种综合性能很好的CBR检索算法.思路是把时间序列相似度比较转化成一个卷积问题，并用DFT来简化这个卷积的计算.通过对这种CBR检索算法进行了深入的理论分析和认真的实验，结果证明，提出的算法是一个高效的算法.在这个检索算法的基础上，CBR就能够应用到时序数据的分析推理中，具有广阔的应用前景.

Abstract:: This paper focuses on the retrieval algorithms of a special kind of CBR system i n which cases are composed of timeseries data. We introduced the classical alg o rithm used for processing similarity queries on time series data. This algorithm is based on the fact that DFT preserves the Euclidean distance in the time or f requency domain, and only the first few elements of the frequency sequence are s ignificant, so the retrieval process can only use these significant elements to compute similarity degree. However, this algorithm has several disadvantages lim iting its usage in CBR retrieval, so a new algorithm is presented for using batc h meth od to compute the similarity degree. It is based on the observation that the ori ginal problem can be transformed to a convolution problem, and the circular conv olution can be computed more efficiently using FFT. Theoretical analysis and exp eriment result prove that this algorithm is efficient and robust. The algorithm presented in this paper furnishes the CBR with the ability to process cases cons ist of timeseries data.

参考文献/References:: ［1］AAMODT A,PLAZA E. Casebased reasoning: foundational issues, metho dological variations, and system approaches［J］. AI Communications, 1994(7):56- 72.
［2］MA J, KNIGHT B. A Framework for Historical Case-Based Reasoning［A］. ICCBR 2003［C］. Trondheim, Norway, 2003.
［3］JAERE M D, AAMODT A, SKALLE P. Representing temporal knowledge for case based prediction［A］. ECCBR 2002［C］. Aberdeen, Scotland, UK, 2002.
［4］ALLEN J F. Maintaining knowledge about temporal intervals［J］. Communi cations of the ACM 1983, 26(11): 832-843.
［5］AGRAWAL R, FALOUTSOS C,SWAMI A. Efficient similarity search in sequence database［A］. FODO Conference［C］. Evanston, Illinois, 1993.
［6］FALOUTSOS C, RANGANATHAN M, MANOLOPOULOS Y. Fast subsequence matching in timeseries database［A］. Proc of the ACM SIGMOD［C］. Minneapolis, Minnesot a, 1994.
［7］GUTTMAN A. Rtrees: a dynamic index structure for spatial searching［A］. Proceedings of the ACM SIGMOD［C］. Boston, MA, 1984.
［8］BERCHTOLD S., BOHM C, KRIEGEL H. The pyramidtechnique: towards breaking t he curse of dimensionality［A］. Proceedings of SIGMOD’98［C］. Seattle, Washin gton, USA, 1998.
［9］FALOUTSOS C,ROSEMAN S. Fractals for secondary key retrieval［R］. Technical Report UMIACSTR8947, CSTR2242, University of Maryland, College Park, Maryland, 1989.
［10］BOHM C, BERCHTOLD S,KEIM D. Searching in high-dimensional spacesindex st ructures for improving the performance of multimedia databases［J］. ACM Computi ng Surveys,2001, 33(3): 322-373.
［11］GAEDE V, GUNTHER O. Multidimensional access method［J］. ACM Computing Surveys, 1998, 30(2):221-290.
［12］BRACEWELL R. The fourier transform and its applications［M］. McGrawHi ll, 2000.
［13］史忠植.高级人工智能:第二版［M］. 北京：科学出版社, 2006.
［14］HAYKIN S, 叶世伟, 史忠植.神经网络原理［M］.北京：机械工业出版社, 2004
［15］吴镇扬.数字信号处理［M］.北京：高等教育出版社， 2004.
［16］MONTANI S，PORTINALE L. Case based representation and retrieval with tim e dependent features［A］. ICCBR 2005［C］. Chicago, IL, USA, 2005.
［17］SANCHEZ M M, CORTES U. An approach for temporal casebased reasoning: ep isodebased reasoning［A］. ICCBR 2005［C］. Chicago, IL, USA, 2005.

备注/Memo

收稿日期：2006-07-10.
基金项目：国家自然科学基金资助项目(60435010，90604017，60675010)；国家“973”资助项目(2003 CB317004)
作者简介：
史忠植, 男，1941年生，中国科学院计算所主任研究员，博士生导师.IEEE高级会员.主要研究领域为智能科学、分布智能、机器学习、知识工程等.1979年、1998年、2001年均获中国科学院科技进步二等奖，1994年获中国科学院科技进步特等奖，2002年获国家科技进步二等奖 .E-mail：shizz@ics.ict.ac.cn.
尹超，男,1979年生，硕士研究生,主要研究方向为智能信息处理,基于范例的推理技术. 
E-mail:yinchao04@mails.gucas.ac.cn.

更新日期/Last Update: 2009-05-05

一种支持时间序列数据的CBR检索算法 PDF下载HTML

备注/Memo

一种支持时间序列数据的CBR检索算法

PDF下载 HTML