Predicting the future performance of soccer players

Publikation: Beiträge in ZeitschriftenZeitschriftenaufsätzeForschungbegutachtet

Standard

Predicting the future performance of soccer players. / Arndt, Cornelius; Brefeld, Ulf.
in: Statistical Analysis and Data Mining, Jahrgang 9, Nr. 5, 01.10.2016, S. 373-382.

Publikation: Beiträge in ZeitschriftenZeitschriftenaufsätzeForschungbegutachtet

Harvard

APA

Vancouver

Bibtex

@article{fe71904951b5445db7e91a2df5d7812f,
title = "Predicting the future performance of soccer players",
abstract = "We propose a multitask, regression-based approach for predicting future performances of soccer players. The multitask approach allows us to simultaneously learn individual player models as offsets to a general model. We devise multitask variants of ridge regression and ε-support vector regression. Together with a hashed joint feature space, the generalized models can be optimized using standard techniques. Relevant features for the prediction are identified by a modified recursive feature elimination strategy. We report on extensive empirical results using real data from the German Bundesliga.",
keywords = "feature selection, machine learning, multitask regression, predictive analytics, ridge regression, support vector regression, Engineering",
author = "Cornelius Arndt and Ulf Brefeld",
year = "2016",
month = oct,
day = "1",
doi = "10.1002/sam.11321",
language = "English",
volume = "9",
pages = "373--382",
journal = "Statistical Analysis and Data Mining",
issn = "1932-1864",
publisher = "John Wiley & Sons Inc.",
number = "5",

}

RIS

TY - JOUR

T1 - Predicting the future performance of soccer players

AU - Arndt, Cornelius

AU - Brefeld, Ulf

PY - 2016/10/1

Y1 - 2016/10/1

N2 - We propose a multitask, regression-based approach for predicting future performances of soccer players. The multitask approach allows us to simultaneously learn individual player models as offsets to a general model. We devise multitask variants of ridge regression and ε-support vector regression. Together with a hashed joint feature space, the generalized models can be optimized using standard techniques. Relevant features for the prediction are identified by a modified recursive feature elimination strategy. We report on extensive empirical results using real data from the German Bundesliga.

AB - We propose a multitask, regression-based approach for predicting future performances of soccer players. The multitask approach allows us to simultaneously learn individual player models as offsets to a general model. We devise multitask variants of ridge regression and ε-support vector regression. Together with a hashed joint feature space, the generalized models can be optimized using standard techniques. Relevant features for the prediction are identified by a modified recursive feature elimination strategy. We report on extensive empirical results using real data from the German Bundesliga.

KW - feature selection

KW - machine learning

KW - multitask regression

KW - predictive analytics

KW - ridge regression

KW - support vector regression

KW - Engineering

UR - http://www.scopus.com/inward/record.url?scp=84988431971&partnerID=8YFLogxK

U2 - 10.1002/sam.11321

DO - 10.1002/sam.11321

M3 - Journal articles

AN - SCOPUS:84988431971

VL - 9

SP - 373

EP - 382

JO - Statistical Analysis and Data Mining

JF - Statistical Analysis and Data Mining

SN - 1932-1864

IS - 5

ER -

DOI