Streaming pattern discovery in multiple time-series

Spiros Papadimitriou, Jimeng Sun, Christos Faloutsos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we introduce SPIRIT (Streaming Pattern dIscoveRy in multIple Time-series). Given n numerical data streams, all of whose values we observe at each time tick t, SPIRIT can incrementally nd correlations and hidden variables, which summarise the key trends in the entire stream collection. It can do this quickly, with no bu ering of stream values and without comparing pairs of streams. Moreover, it is any-time, single pass, and it dynamically detects changes. The discovered trends can also be used to immediately spot potential anomalies, to do e cient forecasting and, more generally, to dramatically simplify further data processing. Our experimental evaluation and case studies show that SPIRIT can incrementally capture correlations and discover trends, e ciently and e ectively.

Original languageEnglish (US)
Title of host publicationVLDB 2005 - Proceedings of 31st International Conference on Very Large Data Bases
Pages697-708
Number of pages12
StatePublished - Dec 1 2005
Externally publishedYes
EventVLDB 2005 - 31st International Conference on Very Large Data Bases - Trondheim, Norway
Duration: Aug 30 2005Sep 2 2005

Publication series

NameVLDB 2005 - Proceedings of 31st International Conference on Very Large Data Bases
Volume2

Other

OtherVLDB 2005 - 31st International Conference on Very Large Data Bases
CountryNorway
CityTrondheim
Period8/30/059/2/05

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'Streaming pattern discovery in multiple time-series'. Together they form a unique fingerprint.

  • Cite this

    Papadimitriou, S., Sun, J., & Faloutsos, C. (2005). Streaming pattern discovery in multiple time-series. In VLDB 2005 - Proceedings of 31st International Conference on Very Large Data Bases (pp. 697-708). (VLDB 2005 - Proceedings of 31st International Conference on Very Large Data Bases; Vol. 2).