An incremental anytime algorithm for mining T-patterns from event streams

Keith Johnson, Wei Liu

    Research output: Chapter in Book/Conference paperConference paper

    Abstract

    Temporal patterns that capture frequent time differences occurring between items in a sequence are gaining increasing attention as a growing research area. Time-interval sequential patterns (also known as T-Patterns) not only capture the order of symbols but also the time delay between symbols, where the time delay is specified as a time-interval between a pair of symbols. Such patterns have been shown to be present in many different types of data (e.g. spike data, smart home activity, DNA sequences, human and animal behaviour analysis and the like) which cannot be captured by other pattern types. Recently, several mining algorithms have been proposed to mine such patterns from either transaction databases or static sequences of time-stamped events. However, they are not capable of online mining from streams of time-stamped events (i.e. event streams). An increasingly common form of data, event streams bring more challenges as they are often unsegmented and with unobtainable total size. In this paper, we propose a mining algorithm that discovers time-interval patterns online, from event streams and demonstrate its capability on a benchmark synthetic dataset.

    Original languageEnglish
    Title of host publicationData Mining - 15th Australasian Conference, AusDM 2017, Revised Selected Papers
    EditorsDavid Stirling, Yee Ling Boo, Lianhua Chi, Kok-Leong Ong, Lin Liu, Graham Williams
    PublisherSpringer-Verlag Wien
    Pages144-157
    Number of pages14
    ISBN (Print)9789811302916
    DOIs
    Publication statusPublished - 1 Jan 2018
    Event15th Australasian Conference on Data Mining, AusDM 2017 - Melbourne, Australia
    Duration: 19 Aug 201720 Aug 2017

    Publication series

    NameCommunications in Computer and Information Science
    Volume845
    ISSN (Print)1865-0929

    Conference

    Conference15th Australasian Conference on Data Mining, AusDM 2017
    CountryAustralia
    CityMelbourne
    Period19/08/1720/08/17

    Fingerprint

    Incremental Algorithm
    Mining
    Time delay
    DNA sequences
    Animals
    Interval
    Time Delay
    Sequential Patterns
    Smart Home
    Spike
    DNA Sequence
    Transactions
    Benchmark
    Demonstrate

    Cite this

    Johnson, K., & Liu, W. (2018). An incremental anytime algorithm for mining T-patterns from event streams. In D. Stirling, Y. L. Boo, L. Chi, K-L. Ong, L. Liu, & G. Williams (Eds.), Data Mining - 15th Australasian Conference, AusDM 2017, Revised Selected Papers (pp. 144-157). (Communications in Computer and Information Science; Vol. 845). Springer-Verlag Wien. https://doi.org/10.1007/978-981-13-0292-3_9
    Johnson, Keith ; Liu, Wei. / An incremental anytime algorithm for mining T-patterns from event streams. Data Mining - 15th Australasian Conference, AusDM 2017, Revised Selected Papers. editor / David Stirling ; Yee Ling Boo ; Lianhua Chi ; Kok-Leong Ong ; Lin Liu ; Graham Williams. Springer-Verlag Wien, 2018. pp. 144-157 (Communications in Computer and Information Science).
    @inproceedings{0e9a38a6253244828378efefcbf7b1bd,
    title = "An incremental anytime algorithm for mining T-patterns from event streams",
    abstract = "Temporal patterns that capture frequent time differences occurring between items in a sequence are gaining increasing attention as a growing research area. Time-interval sequential patterns (also known as T-Patterns) not only capture the order of symbols but also the time delay between symbols, where the time delay is specified as a time-interval between a pair of symbols. Such patterns have been shown to be present in many different types of data (e.g. spike data, smart home activity, DNA sequences, human and animal behaviour analysis and the like) which cannot be captured by other pattern types. Recently, several mining algorithms have been proposed to mine such patterns from either transaction databases or static sequences of time-stamped events. However, they are not capable of online mining from streams of time-stamped events (i.e. event streams). An increasingly common form of data, event streams bring more challenges as they are often unsegmented and with unobtainable total size. In this paper, we propose a mining algorithm that discovers time-interval patterns online, from event streams and demonstrate its capability on a benchmark synthetic dataset.",
    author = "Keith Johnson and Wei Liu",
    year = "2018",
    month = "1",
    day = "1",
    doi = "10.1007/978-981-13-0292-3_9",
    language = "English",
    isbn = "9789811302916",
    series = "Communications in Computer and Information Science",
    publisher = "Springer-Verlag Wien",
    pages = "144--157",
    editor = "David Stirling and Boo, {Yee Ling} and Lianhua Chi and Kok-Leong Ong and Lin Liu and Graham Williams",
    booktitle = "Data Mining - 15th Australasian Conference, AusDM 2017, Revised Selected Papers",
    address = "Austria",

    }

    Johnson, K & Liu, W 2018, An incremental anytime algorithm for mining T-patterns from event streams. in D Stirling, YL Boo, L Chi, K-L Ong, L Liu & G Williams (eds), Data Mining - 15th Australasian Conference, AusDM 2017, Revised Selected Papers. Communications in Computer and Information Science, vol. 845, Springer-Verlag Wien, pp. 144-157, 15th Australasian Conference on Data Mining, AusDM 2017, Melbourne, Australia, 19/08/17. https://doi.org/10.1007/978-981-13-0292-3_9

    An incremental anytime algorithm for mining T-patterns from event streams. / Johnson, Keith; Liu, Wei.

    Data Mining - 15th Australasian Conference, AusDM 2017, Revised Selected Papers. ed. / David Stirling; Yee Ling Boo; Lianhua Chi; Kok-Leong Ong; Lin Liu; Graham Williams. Springer-Verlag Wien, 2018. p. 144-157 (Communications in Computer and Information Science; Vol. 845).

    Research output: Chapter in Book/Conference paperConference paper

    TY - GEN

    T1 - An incremental anytime algorithm for mining T-patterns from event streams

    AU - Johnson, Keith

    AU - Liu, Wei

    PY - 2018/1/1

    Y1 - 2018/1/1

    N2 - Temporal patterns that capture frequent time differences occurring between items in a sequence are gaining increasing attention as a growing research area. Time-interval sequential patterns (also known as T-Patterns) not only capture the order of symbols but also the time delay between symbols, where the time delay is specified as a time-interval between a pair of symbols. Such patterns have been shown to be present in many different types of data (e.g. spike data, smart home activity, DNA sequences, human and animal behaviour analysis and the like) which cannot be captured by other pattern types. Recently, several mining algorithms have been proposed to mine such patterns from either transaction databases or static sequences of time-stamped events. However, they are not capable of online mining from streams of time-stamped events (i.e. event streams). An increasingly common form of data, event streams bring more challenges as they are often unsegmented and with unobtainable total size. In this paper, we propose a mining algorithm that discovers time-interval patterns online, from event streams and demonstrate its capability on a benchmark synthetic dataset.

    AB - Temporal patterns that capture frequent time differences occurring between items in a sequence are gaining increasing attention as a growing research area. Time-interval sequential patterns (also known as T-Patterns) not only capture the order of symbols but also the time delay between symbols, where the time delay is specified as a time-interval between a pair of symbols. Such patterns have been shown to be present in many different types of data (e.g. spike data, smart home activity, DNA sequences, human and animal behaviour analysis and the like) which cannot be captured by other pattern types. Recently, several mining algorithms have been proposed to mine such patterns from either transaction databases or static sequences of time-stamped events. However, they are not capable of online mining from streams of time-stamped events (i.e. event streams). An increasingly common form of data, event streams bring more challenges as they are often unsegmented and with unobtainable total size. In this paper, we propose a mining algorithm that discovers time-interval patterns online, from event streams and demonstrate its capability on a benchmark synthetic dataset.

    UR - http://www.scopus.com/inward/record.url?scp=85045843334&partnerID=8YFLogxK

    U2 - 10.1007/978-981-13-0292-3_9

    DO - 10.1007/978-981-13-0292-3_9

    M3 - Conference paper

    SN - 9789811302916

    T3 - Communications in Computer and Information Science

    SP - 144

    EP - 157

    BT - Data Mining - 15th Australasian Conference, AusDM 2017, Revised Selected Papers

    A2 - Stirling, David

    A2 - Boo, Yee Ling

    A2 - Chi, Lianhua

    A2 - Ong, Kok-Leong

    A2 - Liu, Lin

    A2 - Williams, Graham

    PB - Springer-Verlag Wien

    ER -

    Johnson K, Liu W. An incremental anytime algorithm for mining T-patterns from event streams. In Stirling D, Boo YL, Chi L, Ong K-L, Liu L, Williams G, editors, Data Mining - 15th Australasian Conference, AusDM 2017, Revised Selected Papers. Springer-Verlag Wien. 2018. p. 144-157. (Communications in Computer and Information Science). https://doi.org/10.1007/978-981-13-0292-3_9