A representing model of rule distribution in temporal sequence

In recent years, there has been a lot of interest in using data mining techniques to extract rules from temporal sequences in various applications. Previous works on rule discovery mainly considered global pattern behaviors. In this paper, we consider the rules of which the frequency is large only in a subsequence of the original sequence. To facilitate the discovery of rule distribution, we present a representing model, which is to segment the sequence into a set of continuous subsequence, in which there exists a rule set that appears frequently. We present the definition of local rule and our model, together with the relating methods. We analyze the behavior of the problem and our algorithms with both synthetic and real data. The results obtained correspond with the definition of our problem and reveal a kind of novel knowledge.

[1]  Heikki Mannila,et al.  Rule Discovery from Time Series , 1998, KDD.

[2]  Ramakrishnan Srikant,et al.  Mining Sequential Patterns: Generalizations and Performance Improvements , 1996, EDBT.

[3]  Sushil Jajodia,et al.  Discovering Temporal Patterns in Multiple Granularities , 2000, TSDM.

[4]  Hagit Shatkay,et al.  Approximate queries and representations for large data sequences , 1996, Proceedings of the Twelfth International Conference on Data Engineering.

[5]  Padhraic Smyth,et al.  From Data Mining to Knowledge Discovery: An Overview , 1996, Advances in Knowledge Discovery and Data Mining.

[6]  Jaideep Srivastava,et al.  Event detection from time series data , 1999, KDD '99.

[7]  Sunita Sarawagi,et al.  Mining Surprising Patterns Using Temporal Description Length , 1998, VLDB.

[8]  Ada Wai-Chee Fu,et al.  Discovering Temporal Patterns for Interval-Based Events , 2000, DaWaK.

[9]  Ramakrishnan Srikant,et al.  Mining sequential patterns , 1995, Proceedings of the Eleventh International Conference on Data Engineering.