Dominant interaction group detection in team sports

Behzad BOZORGTABAR, Roland GOECKE

    Research output: A Conference proceeding or a Chapter in BookConference contribution

    Abstract

    With sport performance analysis shifting towards a team’s overall behaviour, the need for automatically discovering formations, such as players temporarily forming a common group of interactions, emerges to complement the coach’s observations. We propose a novel framework to detect a structured group of players, encoded in the form of context-dependent team players’ interactions. In real scenarios, considering a tendency among the players’ movements to occupy a common interested region on a sport field, we predict the future candidate area of group interactions (tendentious zone) before the group formations occur. Consequently, the tendentious zone guides the future players’ movements and provides prior information about their future positions on the field. Building a graph of all players’ positions and considering their motion stability towards the tendentious zone, we aim to discover an optimal subgraph indicating a dominant group of players by maximising the similarity among them. To quantify the similarity of any two players, we
    consider their relative proximity as well as the common social attention model. Experiments on new sports datasets consistently show the superiority and effectiveness of the proposed approach over existing group detection methods.
    Original languageEnglish
    Title of host publication21st ACM SIGKDD Conference on knowledge discovery and data mining
    Subtitle of host publicationProceedings of the largescale sports analysis workshop
    EditorsThornsten Joachims, Geoff Webb
    Place of PublicationSydney
    PublisherAssociation for Computing Machinery (ACM)
    Pages1-4
    Number of pages4
    ISBN (Print)9781450336642
    Publication statusPublished - 2015
    Event21st ACM SIGKDD Conference on knowledge discovery and data mining - Sydney, Sydney, Australia
    Duration: 10 Aug 201513 Aug 2015
    http://www.kdd.org/kdd2015/ (Conference detail)

    Conference

    Conference21st ACM SIGKDD Conference on knowledge discovery and data mining
    Abbreviated titleKDD 2015
    CountryAustralia
    CitySydney
    Period10/08/1513/08/15
    Internet address

    Fingerprint

    Sports
    Experiments

    Cite this

    BOZORGTABAR, B., & GOECKE, R. (2015). Dominant interaction group detection in team sports. In T. Joachims, & G. Webb (Eds.), 21st ACM SIGKDD Conference on knowledge discovery and data mining: Proceedings of the largescale sports analysis workshop (pp. 1-4). Sydney: Association for Computing Machinery (ACM).
    BOZORGTABAR, Behzad ; GOECKE, Roland. / Dominant interaction group detection in team sports. 21st ACM SIGKDD Conference on knowledge discovery and data mining: Proceedings of the largescale sports analysis workshop. editor / Thornsten Joachims ; Geoff Webb. Sydney : Association for Computing Machinery (ACM), 2015. pp. 1-4
    @inproceedings{8c9afa6c7ad5448bbfbe6a3d502b63ad,
    title = "Dominant interaction group detection in team sports",
    abstract = "With sport performance analysis shifting towards a team’s overall behaviour, the need for automatically discovering formations, such as players temporarily forming a common group of interactions, emerges to complement the coach’s observations. We propose a novel framework to detect a structured group of players, encoded in the form of context-dependent team players’ interactions. In real scenarios, considering a tendency among the players’ movements to occupy a common interested region on a sport field, we predict the future candidate area of group interactions (tendentious zone) before the group formations occur. Consequently, the tendentious zone guides the future players’ movements and provides prior information about their future positions on the field. Building a graph of all players’ positions and considering their motion stability towards the tendentious zone, we aim to discover an optimal subgraph indicating a dominant group of players by maximising the similarity among them. To quantify the similarity of any two players, weconsider their relative proximity as well as the common social attention model. Experiments on new sports datasets consistently show the superiority and effectiveness of the proposed approach over existing group detection methods.",
    keywords = "interaction group, Team sport",
    author = "Behzad BOZORGTABAR and Roland GOECKE",
    year = "2015",
    language = "English",
    isbn = "9781450336642",
    pages = "1--4",
    editor = "Thornsten Joachims and Geoff Webb",
    booktitle = "21st ACM SIGKDD Conference on knowledge discovery and data mining",
    publisher = "Association for Computing Machinery (ACM)",
    address = "United States",

    }

    BOZORGTABAR, B & GOECKE, R 2015, Dominant interaction group detection in team sports. in T Joachims & G Webb (eds), 21st ACM SIGKDD Conference on knowledge discovery and data mining: Proceedings of the largescale sports analysis workshop. Association for Computing Machinery (ACM), Sydney, pp. 1-4, 21st ACM SIGKDD Conference on knowledge discovery and data mining, Sydney, Australia, 10/08/15.

    Dominant interaction group detection in team sports. / BOZORGTABAR, Behzad; GOECKE, Roland.

    21st ACM SIGKDD Conference on knowledge discovery and data mining: Proceedings of the largescale sports analysis workshop. ed. / Thornsten Joachims; Geoff Webb. Sydney : Association for Computing Machinery (ACM), 2015. p. 1-4.

    Research output: A Conference proceeding or a Chapter in BookConference contribution

    TY - GEN

    T1 - Dominant interaction group detection in team sports

    AU - BOZORGTABAR, Behzad

    AU - GOECKE, Roland

    PY - 2015

    Y1 - 2015

    N2 - With sport performance analysis shifting towards a team’s overall behaviour, the need for automatically discovering formations, such as players temporarily forming a common group of interactions, emerges to complement the coach’s observations. We propose a novel framework to detect a structured group of players, encoded in the form of context-dependent team players’ interactions. In real scenarios, considering a tendency among the players’ movements to occupy a common interested region on a sport field, we predict the future candidate area of group interactions (tendentious zone) before the group formations occur. Consequently, the tendentious zone guides the future players’ movements and provides prior information about their future positions on the field. Building a graph of all players’ positions and considering their motion stability towards the tendentious zone, we aim to discover an optimal subgraph indicating a dominant group of players by maximising the similarity among them. To quantify the similarity of any two players, weconsider their relative proximity as well as the common social attention model. Experiments on new sports datasets consistently show the superiority and effectiveness of the proposed approach over existing group detection methods.

    AB - With sport performance analysis shifting towards a team’s overall behaviour, the need for automatically discovering formations, such as players temporarily forming a common group of interactions, emerges to complement the coach’s observations. We propose a novel framework to detect a structured group of players, encoded in the form of context-dependent team players’ interactions. In real scenarios, considering a tendency among the players’ movements to occupy a common interested region on a sport field, we predict the future candidate area of group interactions (tendentious zone) before the group formations occur. Consequently, the tendentious zone guides the future players’ movements and provides prior information about their future positions on the field. Building a graph of all players’ positions and considering their motion stability towards the tendentious zone, we aim to discover an optimal subgraph indicating a dominant group of players by maximising the similarity among them. To quantify the similarity of any two players, weconsider their relative proximity as well as the common social attention model. Experiments on new sports datasets consistently show the superiority and effectiveness of the proposed approach over existing group detection methods.

    KW - interaction group

    KW - Team sport

    M3 - Conference contribution

    SN - 9781450336642

    SP - 1

    EP - 4

    BT - 21st ACM SIGKDD Conference on knowledge discovery and data mining

    A2 - Joachims, Thornsten

    A2 - Webb, Geoff

    PB - Association for Computing Machinery (ACM)

    CY - Sydney

    ER -

    BOZORGTABAR B, GOECKE R. Dominant interaction group detection in team sports. In Joachims T, Webb G, editors, 21st ACM SIGKDD Conference on knowledge discovery and data mining: Proceedings of the largescale sports analysis workshop. Sydney: Association for Computing Machinery (ACM). 2015. p. 1-4