Learning to Rate Player Positioning in Soccer

Publikation: Beiträge in ZeitschriftenZeitschriftenaufsätzeForschungbegutachtet

Standard

Learning to Rate Player Positioning in Soccer. / Dick, Uwe; Brefeld, Ulf.

in: Big Data, Jahrgang 7, Nr. 1, 01.03.2019, S. 71-82.

Publikation: Beiträge in ZeitschriftenZeitschriftenaufsätzeForschungbegutachtet

Harvard

APA

Vancouver

Dick U, Brefeld U. Learning to Rate Player Positioning in Soccer. Big Data. 2019 Mär 1;7(1):71-82. doi: 10.1089/big.2018.0054

Bibtex

@article{04200661b7d04b428afe977aede22ee4,
title = "Learning to Rate Player Positioning in Soccer",
abstract = "We investigate how to learn functions that rate game situations on a soccer pitch according to their potential to lead to successful attacks. We follow a purely data-driven approach using techniques from deep reinforcement learning to valuate multiplayer positionings based on positional data. Empirically, the predicted scores highly correlate with dangerousness of actual situations and show that rating of player positioning without expert knowledge is possible.",
keywords = "Informatics, deep learning, reinforcement learning, scoring function, spatiotemportal data, Business informatics",
author = "Uwe Dick and Ulf Brefeld",
note = "Publisher Copyright: Copyright 2019, Mary Ann Liebert, Inc., publishers",
year = "2019",
month = mar,
day = "1",
doi = "10.1089/big.2018.0054",
language = "English",
volume = "7",
pages = "71--82",
journal = "Big Data",
issn = "2167-6461",
publisher = "Mary Ann Liebert Inc.",
number = "1",

}

RIS

TY - JOUR

T1 - Learning to Rate Player Positioning in Soccer

AU - Dick, Uwe

AU - Brefeld, Ulf

N1 - Publisher Copyright: Copyright 2019, Mary Ann Liebert, Inc., publishers

PY - 2019/3/1

Y1 - 2019/3/1

N2 - We investigate how to learn functions that rate game situations on a soccer pitch according to their potential to lead to successful attacks. We follow a purely data-driven approach using techniques from deep reinforcement learning to valuate multiplayer positionings based on positional data. Empirically, the predicted scores highly correlate with dangerousness of actual situations and show that rating of player positioning without expert knowledge is possible.

AB - We investigate how to learn functions that rate game situations on a soccer pitch according to their potential to lead to successful attacks. We follow a purely data-driven approach using techniques from deep reinforcement learning to valuate multiplayer positionings based on positional data. Empirically, the predicted scores highly correlate with dangerousness of actual situations and show that rating of player positioning without expert knowledge is possible.

KW - Informatics

KW - deep learning

KW - reinforcement learning

KW - scoring function

KW - spatiotemportal data

KW - Business informatics

UR - http://www.scopus.com/inward/record.url?scp=85063285321&partnerID=8YFLogxK

U2 - 10.1089/big.2018.0054

DO - 10.1089/big.2018.0054

M3 - Journal articles

C2 - 30672712

VL - 7

SP - 71

EP - 82

JO - Big Data

JF - Big Data

SN - 2167-6461

IS - 1

ER -

DOI