Određivanje vrijednosti nogometnih igrača primjenom strojnog učenja

master's thesis

• Title (english): Valuation of football players using machine learning
• Author: Ante Mislav Vukojević
• Mentor: Tomislav Globan (mentor)
• Committee member: Davor Labaš (član povjerenstva)
• Committee member: Vladimir Arčabić (član povjerenstva)
• Committee member: Tomislav Globan (predsjednik povjerenstva)
• Granter: University of Zagreb; Faculty of Economics and Business; (Department of Managerial Economics); Zagreb
• Defense date and country: 2024-10-21, Croatia
• Scientific / art field, discipline and subdiscipline: SOCIAL SCIENCES; Economics; Quantitative Economics

Abstract

Ovaj diplomski rad istražuje primjenu naprednih metoda strojnog učenja u vrednovanju nogometnih igrača, nadovezujući se na postojeća istraživanja koja povezuju ekonomsku teoriju s modernom podatkovnom analitikom. Rad analizira kompleksnost nogometnog tržišta, promatrajući klubove kao subjekte koji maksimiziraju korisnost te ispitujući specifičnosti tržišta rada u nogometu. Razvijen je model temeljen na CatBoost algoritmu, koji koristi širok spektar statističkih i performansnih podataka igrača za predviđanje njihove tržišne vrijednosti. Istraživanje potvrđuje važnost faktora poput snage lige i kluba u vrednovanju igrača te identificira izazove u procjeni vrijednosti vrhunskih igrača i mladih talenata. Rezultati doprinose postojećim alatima za donošenje informiranih odluka o transferima i razvoju igrača. Zaključno, iako model ne može zamijeniti stručnu procjenu, on pruža dodatni, podatkovni pristup koji može unaprijediti proces odlučivanja u upravljanju profesionalnim nogometnim klubovima.

Abstract (english)

This thesis examines the application of advanced machine learning techniques in football player valuation, building upon existing research that combines economic theory with modern data analytics. The study analyzes the complexities of the football market, viewing clubs as utility-maximizing entities and investigating the unique aspects of the football labor market. A model based on the CatBoost algorithm is developed, utilizing a wide range of player statistics and performance data to predict their market values. The research confirms the importance of factors such as league and club strength in player valuation and identifies challenges in estimating the value of top players and young talents. The findings contribute to existing tools for informed decision-making in player transfers and development. In conclusion, while the model cannot replace expert judgment, it provides an additional, data-driven approach that can enhance the decision-making process in professional football club management.

• Keywords: vrednovanje nogometaša; strojno učenje; CatBoost algoritam; nogometni transferi; sportska ekonomija; analitika velikih podataka; nogometni menadžment; Dinamo
• Keywords (english): football player valuation; machine learning; CatBoost algorithm; football transfers; sports economics; big data analytics; football management
• Language: croatian
• URN:NBN: urn:nbn:hr:148:956536
• Study programme: Title: Business economics; Study programme type: university; Study level: integrated undergraduate and graduate; Academic / professional title: sveučilišni magistar ekonomije/sveučilišna magistra ekonomije (sveučilišni magistar ekonomije/sveučilišna magistra ekonomije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Access restricted to students and staff of home institution
• Created on: 2024-10-23 23:48:21