Show simple item record

dc.contributor.authorJauhiainen S.
dc.contributor.authorÄyrämö S.
dc.contributor.authorForsman H.
dc.contributor.authorKauppi J-P.
dc.date.accessioned2020-01-03T09:31:24Z
dc.date.available2020-01-03T09:31:24Z
dc.date.issued2019
dc.identifier.citationJauhiainen S., Äyrämö S., Forsman H., Kauppi J-P. (2019). Talent identification in soccer using a one-class support vector machine. <i>International Journal of Computer Science in Sport</i>, <i>18</i>(3), 125-136. <a href="https://doi.org/10.2478/ijcss-2019-0021" target="_blank">https://doi.org/10.2478/ijcss-2019-0021</a>
dc.identifier.otherCONVID_33915511
dc.identifier.urihttps://jyx.jyu.fi/handle/123456789/67083
dc.description.abstractIdentifying potential future elite athletes is important in many sporting events. The successful identification of potential future elite athletes at an early age would help to provide high-quality coaching and training environments in which to optimize their development. However, a large variety of different skills and qualities are needed to succeed in elite sports, making talent identification generally a complex and multifaceted problem. Due to the rarity of elite athletes, datasets are inherently imbalanced, making classical statistical inference difficult. Therefore, we approach talent identification as an anomaly detection problem. We trained a nonlinear one-class support vector machine (one-class SVM) on a dataset (N=951) collected from 14-year-old junior soccer players to detect potential future elite players. The mean area under the receiver operating characteristic curve (AUC-ROC) over the tested hyperparameter combinations was 0.763 (std 0.007). The most accurate model was obtained when physical tests, measuring, for example, technical skills, speed, and agility, were used. According to our results, the proposed approach could be useful to support decision-makers in the process of talent identification.en
dc.format.mimetypeapplication/pdf
dc.languageeng
dc.language.isoeng
dc.publisherSciendo
dc.relation.ispartofseriesInternational Journal of Computer Science in Sport
dc.rightsCC BY-NC-ND 4.0
dc.subject.othertalent identification
dc.subject.otheranomaly detection
dc.subject.otherone-class svm
dc.titleTalent identification in soccer using a one-class support vector machine
dc.typearticle
dc.identifier.urnURN:NBN:fi:jyu-202001031017
dc.contributor.laitosInformaatioteknologian tiedekuntafi
dc.contributor.laitosFaculty of Information Technologyen
dc.type.urihttp://purl.org/eprint/type/JournalArticle
dc.description.reviewstatuspeerReviewed
dc.format.pagerange125-136
dc.relation.issn1684-4769
dc.relation.numberinseries3
dc.relation.volume18
dc.type.versionpublishedVersion
dc.rights.copyright© The Authors 2019
dc.rights.accesslevelopenAccessfi
dc.relation.grantnumber286019
dc.subject.ysolahjakkuus
dc.subject.ysojalkapallo
dc.subject.ysokoneoppiminen
dc.subject.ysotiedonlouhinta
dc.subject.ysolajitaidot
dc.subject.ysotunnistaminen
dc.format.contentfulltext
jyx.subject.urihttp://www.yso.fi/onto/yso/p6729
jyx.subject.urihttp://www.yso.fi/onto/yso/p6409
jyx.subject.urihttp://www.yso.fi/onto/yso/p21846
jyx.subject.urihttp://www.yso.fi/onto/yso/p5520
jyx.subject.urihttp://www.yso.fi/onto/yso/p25205
jyx.subject.urihttp://www.yso.fi/onto/yso/p8265
dc.rights.urlhttps://creativecommons.org/licenses/by-nc-nd/4.0/
dc.relation.doi10.2478/ijcss-2019-0021
dc.relation.funderSuomen Akatemiafi
dc.relation.funderAcademy of Finlanden
jyx.fundingprogramTutkijatohtori, SAfi
jyx.fundingprogramPostdoctoral Researcher, AoFen
jyx.fundinginformationThis work has been carried out in two projects "Value from health data with cognitive computing" and "Watson Health Cloud", funded by Business Finland. Jukka-Pekka Kauppi was funded by the Academy of Finland Postdoctoral Researcher program (Research Council for Natural Sciences and Engineering; grant number 286019).


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

CC BY-NC-ND 4.0
Except where otherwise noted, this item's license is described as CC BY-NC-ND 4.0