GR Semicolon EN

Show simple item record

dc.contributor.author
Sarlis, Vangelis
en
dc.date.accessioned
2024-05-21T11:56:56Z
dc.date.available
2024-05-21T11:56:56Z
dc.date.issued
2024-05-21
dc.identifier.uri
https://repository.ihu.edu.gr//xmlui/handle/11544/30399
dc.rights
Default License
dc.subject
Data Science
en
dc.subject
Machine Learning
en
dc.subject
Sports Analytics
en
dc.subject
Data Mining
en
dc.subject
Artificial Intelligence
en
dc.subject
Injury Analytics
en
dc.subject
Sports Economics
en
dc.subject
Basketball Analytics
en
dc.subject
Business Intelligence
en
dc.subject
Data Analysis
en
dc.subject
Musculoskeletal Injuries
en
dc.subject
Statistics
en
dc.subject
Sports Injuries
en
dc.subject
Text Analytics
en
dc.subject
Text Mining
en
dc.title
Data Science for Sports Analytics
en
heal.type
doctoralThesis
en_US
heal.creatorID.email
e.sarlis@ihu.edu.gr
heal.generalDescription
Phd Thesis in Data Science for Sports Analytics
en
heal.dateAvailable
2024-05-21
heal.language
en
en_US
heal.access
free
en_US
heal.license
http://creativecommons.org/licenses/by-nc/4.0
en_US
heal.fileFormat
pdf
en_US
heal.recordProvider
School of Science and Technology, PhD
en_US
heal.publicationDate
2024-01
heal.bibliographicCitation
Sarlis, Vangelis, "Data Science for Sports Analytics", PhD thesis, School of Science & Technology, International Hellenic University, 2024
en
heal.abstract
This thesis ventures into the realm of Data Science (DS), and Sports Analytics are becoming pivotal in shaping the future of professional sports. This thesis presents a comprehensive exploration of their application within professional basketball, particularly focusing on the National Basketball Association (NBA). The primary aim of this research is to employ Machine Learning (ML) and Data Mining (DM) techniques to deepen the understanding of player performance, injury patterns, and economic impacts, thereby enabling more informed decision-making in sports management. The objectives of this research are manifold. First, it seeks to benchmark existing performance analytics and propose advanced algorithmic models to enhance the predictive power and understanding of player and team performance metrics in basketball. This involves a detailed analysis of NBA data spanning from 1996 to 2023, providing a longitudinal perspective on the evolution of the game and its players. Second, the study aims to quantify the relationship between player injuries and performance by examining how demographic factors such as age and position, as well as socioeconomic aspects, affect this dynamic. A particular focus is placed on musculoskeletal injuries, their prevalence, and the implications for player career trajectories and team strategies. An additional objective is to analyse the economic ramifications of injuries, identifying the costliest types and their impacts on team finances and player income. This research aims to provide actionable insights for coaches, managers, and healthcare professionals to optimize player care, team composition, and performance strategies. This study employs a variety of ML and DS techniques, including feature selection, clustering, and classification methods, to achieve these aims. Through comprehensive data analysis, it was benchmarked and proposed new algorithmic models to enhance the understanding and prediction of performance metrics. This study delves into the correlation of injuries with players' age, position, and performance, aiming to identify patterns and quantify their financial and strategic impacts. We explore the socioeconomic and demographic factors influencing sports, offering insights into the most prevalent injuries and their implications for the game. Another aim is to focus on understanding injury patterns in the NBA and how these injuries impact player performance. Utilizing a unique dataset, it identifies prevalent injuries, the anatomical areas most affected, and explores the influence of these injuries on players' performance post-recovery. The study stands out for its integrative method, merging injury data with performance and salary information to shed light on the interconnections between injuries, economic impacts, and on-court performance. It also looks into the timing and seasonal nature of injuries to find patterns related to time and external factors, as well as the specific effects of injuries on players' game-by-game performance metrics. This research is aimed at aiding coaches, sports medicine professionals, and team management by providing insights for injury prevention, player rotation optimization, and targeted rehabilitation strategies. The results of this research contribute to a more nuanced understanding of the multifaceted nature of sports performance, the strategic importance of injury management and prevention, and the significant economic considerations involved. By revealing the intricate interplay between these elements, this thesis offers a valuable resource for sports professionals, decision makers, and researchers, paving the way for more effective strategies in team management and player development. This study also sets a foundation for future work in this area, suggesting new avenues for research and application in the burgeoning field of sports analytics. The findings indicate a nuanced relationship between injuries, player performance, and economic aspects, shedding light on the critical age range for peak performance and the substantial financial burden injuries pose to teams. This research contributes to the fields of sports analytics and data science by providing a deeper understanding of game dynamics and presenting strategies for injury prevention and management, team composition, and performance optimization. This thesis not only aids decision makers in the sports industry but also sets the stage for future research in advanced sports analytics and injury management strategies.
en
heal.tableOfContents
Dedications 2 Acknowledgements 3 Declaration 4 Abstract 5 Keywords 7 Statement of Original Authorship 8 Copyright 9 Consulting Committee 11 Table of Contents 12 List of Tables 15 List of Figures 18 List of Research Questions 19 1. Introduction 20 2. Background 29 2.1 Sports Analytics 29 2.2 Data Mining and Machine Learning Used in Sports 33 2.3 Implementations of Machine Learning and Data Mining Techniques in Sports 35 2.4 Player Injuries in Sports and Data Analysis Techniques 38 2.5 Major injuries that influence performance in the NBA League 39 2.6 Socioeconomic Influence 42 2.7 Sports Analytics and Text Mining NBA Data to Assess Injury Recovery and Economic Impact 44 2.8 Injury Patterns and Impact on Performance in the NBA League Using Sports Analytics 48 3. Data and Methods 53 3.1 Research Questions 54 3.2 Aim and Objectives 58 3.3 Methodology 62 3.3.1 Data Collection 64 3.3.2 Data Engineering 68 3.3.3 Data Analysis 69 4. Results 72 4.1 Sports Analytics – Evaluation of Basketball Players and Team Performance 73 4.2 Sport injuries in the NBA League for 2010-2020 76 4.3 Results through Data Mining and Machine Learning methods used for Sports Injury Analytics 84 4.4 Findings on the Economic and Performance Impact of Injuries, Age and Position on NBA Players 88 4.5 Results in Recovery from Injuries and Their Economic Impact 89 4.6 Results of Injury Patterns and Impact on Performance in the NBA League Using Sports Analytics 104 5. Discussion & Implications 109 5.1 Discussion of Basketball Performance Evaluation 110 5.2 Basketball Performance Evaluation - Case Study 118 5.3 Aggregated Performance Indicator - Forecasting Scenario 121 5.4 Health and Injury Analytics Discussion 126 5.5 Discussion of Socioeconomic and Health Analytics 130 5.6 Injury Recovery and Economic Impact 139 5.7 Discussion on Injury Patterns and Impact on Performance in the NBA League Using Sports Analytics 147 6. Conclusions 149 6.1 Evaluation of Basketball Players and Team Performance 150 6.2 Impact of Injuries on Basketball Player and Team Performance 152 6.3 Economic and Performance Impact of Injuries, Age and Position on NBA Players 154 6.4 Economic Impact of Injuries and Recovery Assessment 155 6.5 Injury Patterns and Impact on Performance in the NBA League Using Sports Analytics 158 7. Future Work 160 7.1 Advanced predictive and prescriptive Sports Analytics approaches 163 7.2 GPS, biometric, and wearable sensor data analysis 163 7.3 Tactics, Strategy and Technical analysis 164 7.4 Health, nutrition, and injury implications 164 7.5 Video data analysis 165 7.6 Trips, workload, sleep, and fatigue correlation 165 7.7 Social network analysis 166 7.8 Budget control, investments, Risk Management, and forecasting analysis 167 7.9 Leadership and clutch skills 167 Appendices 168 Abbreviations 189 References 191
en
heal.advisorName
Tjortjis, Christos
en
heal.committeeMemberName
Evangelidis, Georgios
en
heal.committeeMemberName
Lykas, Aristeidis
en
heal.committeeMemberName
Tjortjis, Christos
en
heal.academicPublisher
School of Science and Technology
en
heal.academicPublisherID
ihu
en_US
heal.numberOfPages
212
en_US
heal.spatialCoverage
USA
en
heal.temporalCoverage
1996-2023
en


This item appears in the following Collection(s)

Show simple item record

Related Items