Sublinear methods for detecting periodic trends in data streams

Funda Ergun, Shanmugavelayutham Muthukrishnan, S. Cenk Sahinalp

    Research output: Contribution to journalArticle

    Abstract

    We present sublinear algorithms -algorithms that use significantly less resources than needed to store or process the entire input stream - for discovering representative trends in data streams in the form of periodicities. Our algorithms involve sampling Õ(√n) positions. and thus they scan not the entire data stream but merely a sublinear sample thereof. Alternately, our algorithms may be thought of as working on streaming inputs where each data item is seen once, but we store only a sublinear - Õ(√n) - size sample from which we can identify periodicities. In this work we present a variety of definitions of periodicities of a given stream, present sublinear sampling algorithms for discovering them, and prove that the algorithms meet our specifications and guarantees. No previously known results can provide such guarantees for finding any such periodic trends. We also investigate the relationships between these different definitions of periodicity.

    Original languageEnglish (US)
    Pages (from-to)16-28
    Number of pages13
    JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume2976
    StatePublished - Dec 1 2004

    Fingerprint

    Data Streams
    Periodicity
    Entire
    Sampling
    Streaming
    Trends
    Sample Size
    Specification
    Specifications
    Resources

    ASJC Scopus subject areas

    • Theoretical Computer Science
    • Computer Science(all)

    Cite this

    Sublinear methods for detecting periodic trends in data streams. / Ergun, Funda; Muthukrishnan, Shanmugavelayutham; Cenk Sahinalp, S.

    In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 2976, 01.12.2004, p. 16-28.

    Research output: Contribution to journalArticle

    @article{b9d8147e5f1f4c7a852a9357186a892b,
    title = "Sublinear methods for detecting periodic trends in data streams",
    abstract = "We present sublinear algorithms -algorithms that use significantly less resources than needed to store or process the entire input stream - for discovering representative trends in data streams in the form of periodicities. Our algorithms involve sampling {\~O}(√n) positions. and thus they scan not the entire data stream but merely a sublinear sample thereof. Alternately, our algorithms may be thought of as working on streaming inputs where each data item is seen once, but we store only a sublinear - {\~O}(√n) - size sample from which we can identify periodicities. In this work we present a variety of definitions of periodicities of a given stream, present sublinear sampling algorithms for discovering them, and prove that the algorithms meet our specifications and guarantees. No previously known results can provide such guarantees for finding any such periodic trends. We also investigate the relationships between these different definitions of periodicity.",
    author = "Funda Ergun and Shanmugavelayutham Muthukrishnan and {Cenk Sahinalp}, S.",
    year = "2004",
    month = "12",
    day = "1",
    language = "English (US)",
    volume = "2976",
    pages = "16--28",
    journal = "Lecture Notes in Computer Science",
    issn = "0302-9743",
    publisher = "Springer Verlag",

    }

    TY - JOUR

    T1 - Sublinear methods for detecting periodic trends in data streams

    AU - Ergun, Funda

    AU - Muthukrishnan, Shanmugavelayutham

    AU - Cenk Sahinalp, S.

    PY - 2004/12/1

    Y1 - 2004/12/1

    N2 - We present sublinear algorithms -algorithms that use significantly less resources than needed to store or process the entire input stream - for discovering representative trends in data streams in the form of periodicities. Our algorithms involve sampling Õ(√n) positions. and thus they scan not the entire data stream but merely a sublinear sample thereof. Alternately, our algorithms may be thought of as working on streaming inputs where each data item is seen once, but we store only a sublinear - Õ(√n) - size sample from which we can identify periodicities. In this work we present a variety of definitions of periodicities of a given stream, present sublinear sampling algorithms for discovering them, and prove that the algorithms meet our specifications and guarantees. No previously known results can provide such guarantees for finding any such periodic trends. We also investigate the relationships between these different definitions of periodicity.

    AB - We present sublinear algorithms -algorithms that use significantly less resources than needed to store or process the entire input stream - for discovering representative trends in data streams in the form of periodicities. Our algorithms involve sampling Õ(√n) positions. and thus they scan not the entire data stream but merely a sublinear sample thereof. Alternately, our algorithms may be thought of as working on streaming inputs where each data item is seen once, but we store only a sublinear - Õ(√n) - size sample from which we can identify periodicities. In this work we present a variety of definitions of periodicities of a given stream, present sublinear sampling algorithms for discovering them, and prove that the algorithms meet our specifications and guarantees. No previously known results can provide such guarantees for finding any such periodic trends. We also investigate the relationships between these different definitions of periodicity.

    UR - http://www.scopus.com/inward/record.url?scp=35048856469&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=35048856469&partnerID=8YFLogxK

    M3 - Article

    AN - SCOPUS:35048856469

    VL - 2976

    SP - 16

    EP - 28

    JO - Lecture Notes in Computer Science

    JF - Lecture Notes in Computer Science

    SN - 0302-9743

    ER -