A fast algorithm for finding frequent episodes in event streams

ABSTRACT

Frequent episode discovery is a popular framework for mining data available as a long sequence of events. An episode is essentially a short ordered sequence of event types and the frequency of an episode is some suitable measure of how often the episode occurs in the data sequence. Recently,we proposed a new frequency measure for episodes based on the notion of non-overlapped occurrences of episodes in the event sequence, and showed that, such a definition, in addition to yielding computationally efficient algorithms, has some important theoretical properties in connecting frequent episode discovery with HMM learning. This paper presents some new algorithms for frequent episode discovery under this non-overlapped occurrences-based frequency definition. The algorithms presented here are better (by a factor of N, where N denotes the size of episodes being discovered) in terms of both time and space complexities when compared to existing methods for frequent episode discovery. We show through some simulation experiments, that our algorithms are very efficient. The new algorithms presented here have arguably the least possible orders of spaceand time complexities for the task of frequent episode discovery.

REFERENCES

• 1: Mikhail Atallah , Robert Gwadera , Wojciech Szpankowski, Detection of Significant Sets of Episodes in Event Sequences, Proceedings of the Fourth IEEE International Conference on Data Mining, p.3-10, November 01-04, 2004
• 2: G. Casas-Garriga. Discovering unbounded episodes in sequential data. In Proceedings of the 7th European Conference on Principles and Practice of Knowledge Discovery in Databases (PKDD'03), pages 83--94, Cavtat-Dubvrovnik, Croatia, 2003.
• 3: Robert Gwadera , Mikhail Atallah , Wojciech Szpankowski, Reliable Detection of Episodes in Event Sequences, Proceedings of the Third IEEE International Conference on Data Mining, p.67, November 19-22, 2003
• 4: R. Gwadera, M. J. Atallah, and W. Szpankowski. Markov models for identification of significant episodes. In Proceedings of the 2005 SIAM International Conference on Data Mining (SDM-05), Newport Beach, California, April 2005.
• 5: S. Laxman. Discovering frequent episodes in event streams: Fast algorithms, connections with HMMs and generalizations. PhD thesis, Dept. of Electrical Engineering, Indian Institute of Science, Bangalore, India, Mar. 2006.
• 6: Srivatsan Laxman , P. S. Sastry , K. P. Unnikrishnan, Discovering Frequent Episodes and Learning Hidden Markov Models: A Formal Connection, IEEE Transactions on Knowledge and Data Engineering, v.17 n.11, p.1505-1517, November 2005 [doi>10.1109/TKDE.2005.181]
• 7: S. Laxman, P. S. Sastry, and K. P. Unnikrishnan. Discovering frequent generalized episodes when events persist for different durations. To appear in IEEE Transactions on Knowledge and Data Engineering, 2007.
• 8: Heikki Mannila , Hannu Toivonen , A. Inkeri Verkamo, Discovery of Frequent Episodes in Event Sequences, Data Mining and Knowledge Discovery, v.1 n.3, p.259-289, 1997 [doi>10.1023/A:1009748302351]
• 9: Nicolas Méger , Christophe Rigotti, Constraint-based mining of episode rules and optimal window sizes, Proceedings of the 8th European Conference on Principles and Practice of Knowledge Discovery in Databases, p.313-324, September 20-24, 2004, Pisa, Italy
• 10: M. Regnier and W. Szpankowski. On pattern frequency occurrences in a Markovian sequence. Algorithmica, 22:631--649, 1998.

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.2 DATABASE MANAGEMENT
      H.2.8 Database applications
          Subjects: Data mining

General Terms:
Algorithms

Keywords:
event streams, frequent episodes, non-overlapped occurrences, temporal data mining