
Machine Learning and Data Mining for Sports Analytics
Description
Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.
The 12 full papers presented together with 4 challenge papers were carefully reviewed and selected from 24 submissions. The papers present a variety of topics, covering the team sports American football, basketball, ice hockey, and soccer, as well as the individual sports cycling and martial arts. In addition, four challenge papers are included, reporting on how to predict pass receivers in soccer.
More details
Other editions
Additional editions

Content
- Intro
- Preface
- Organization
- Contents
- Soccer
- Measuring Football Players' On-the-Ball Contributions from Passes During Games
- 1 Introduction
- 2 Dataset
- 3 Approach
- 3.1 Constructing Possession Sequences
- 3.2 Labeling Possession Sequences
- 3.3 Valuing Passes
- 3.4 Rating Players
- 4 Experimental Evaluation
- 4.1 Datasets
- 4.2 Methodology
- 4.3 Impact of the Parameters
- 4.4 Results
- 4.5 Application: Replacing Manuel Lanzini
- 5 Related Work
- 6 Conclusion
- References
- Forecasting the FIFA World Cup - Combining Result- and Goal-Based Team Ability Parameters
- 1 Introduction
- 2 Related Work
- 3 The Models
- 3.1 The Elo Rating System
- 3.2 Offense-Defense Ratings
- 3.3 Match Result Predictions
- 4 Evaluation Procedures
- 5 Validation on Previous World Cups
- 6 Validation on Domestic League Football
- 7 Conclusion
- References
- Distinguishing Between Roles of Football Players in Play-by-Play Match Event Data
- 1 Introduction
- 2 Dataset
- 3 Player Roles
- 4 Approach
- 4.1 Feature Engineering
- 4.2 Probabilistic Classification
- 5 Experimental Evaluation
- 5.1 Methodology
- 5.2 Discussion of Results
- 6 Related Work
- 7 Conclusion
- References
- Player Valuation in European Football
- 1 Introduction
- 2 Related Work
- 3 Data Collection and Preparation
- 3.1 Data Collection
- 3.2 Data Preparation
- 4 Feature Selection
- 4.1 Filter Method
- 4.2 Wrapper Method
- 4.3 Discussion
- 5 Prediction
- 5.1 Methods
- 5.2 Results
- 5.3 Discussion
- 6 Conclusion
- References
- Ranking the Teams in European Football Leagues with Agony
- 1 Introduction
- 1.1 Our Contributions
- 2 Related Work
- 3 A Partial Order for a Partial Season
- 4 Agony for Sport Leagues with Draws
- 4.1 Agony for Football Leagues
- 5 Evaluation
- 5.1 Case Study on European Football Leagues
- 6 Conclusion
- References
- US Team Sports
- Interpreting Deep Sports Analytics: Valuing Actions and Players in the NHL
- 1 Introduction
- 2 Related Work
- 3 Play Dynamics in NHL
- 4 Q-Values and Action Impact
- 4.1 Compute Q Functions with Deep Reinforcement Learning
- 4.2 Evaluate Players with Impact Metric
- 5 Mimicking DRL with Regression Tree
- 6 Interpreting Q Functions and Impact with Mimic Tree
- 6.1 Compute Feature Importance
- 6.2 Draw Partial Dependence Plot
- 7 Highlighting Exceptional Players
- 8 Conclusion and Future Work
- References
- Player Pairs Valuation in Ice Hockey
- 1 Introduction
- 2 Related Work
- 3 Background
- 4 Player Pair Metrics
- 5 Data-Driven Analysis
- 5.1 Top Pairings
- 5.2 TOI-Based Analysis
- 5.3 Relative Ice Time Together
- 6 Conclusion
- References
- Model Trees for Identifying Exceptional Players in the NHL and NBA Drafts
- 1 Introduction
- 2 Related Work
- 3 Datasets
- 4 Methodology
- 4.1 Success Metrics
- 4.2 Model Trees
- 5 Results
- 5.1 Modelling Results of the NHL Draft
- 5.2 Modelling Results of the NBA Draft
- 6 Case Studies: Exceptional Players and Their Strong Points
- 6.1 Explaining the Rankings: Identify Strong Points
- 6.2 Case Studies
- 7 Conclusion
- References
- Evaluating NFL Plays: Expected Points Adjusted for Schedule
- 1 Introduction
- 2 Adjusting for Opponent Strength
- 2.1 What Is the Impact of the Value Adjustment?
- 3 Case Studies
- 4 Discussion and Conclusions
- Appendix A Intuition Behind the WCS Expected Points Model
- Appendix B Other Adjustment Approaches
- Appendix C 33 Points Equal 1 Win in NFL
- References
- Individual Sports
- Real-Time Power Performance Prediction in Tour de France
- 1 Introduction
- 2 Related Work
- 3 Dataset
- 4 Methodology
- 4.1 Feature Engineering
- 4.2 Regression Model
- 5 Result
- 5.1 Evaluation on Feature Engineering
- 5.2 Evaluation on Regression Models
- 5.3 Qualitative Analysis - Real Deployment in Tour de France 2017
- 6 Conclusion
- References
- Automatic Classification of Strike Techniques Using Limb Trajectory Data
- 1 Introduction
- 2 Methods
- 2.1 Data Collection
- 2.2 Trajectory Extraction and Labeling
- 2.3 Data Cleaning and Analysis
- 2.4 Classification Models
- 3 Results
- 3.1 Limb and Technique Classification
- 3.2 Minimal Required Training Set
- 3.3 Hierarchical Classification
- 3.4 Classification of Skill Level
- 4 Discussion
- 4.1 Applications
- 4.2 Limitations and Further Developments
- References
- Challenge Papers
- Predicting Pass Receiver in Football Using Distance Based Features
- 1 Introduction
- 2 Dataset Presentation
- 2.1 Dataset Exploration
- 3 Model Description
- 3.1 General Approach
- 3.2 Learning Pass Probabilities
- 4 Results
- 4.1 Learning from Failures
- 4.2 Qualitative Analysis
- 5 Conclusion
- References
- Football Pass Prediction Using Player Locations
- 1 Introduction
- 2 Observations About the Data
- 3 The FPP Model
- 4 Experimental Evaluation
- 5 Conclusion
- References
- Deep Learning from Spatial Relations for Soccer Pass Prediction
- 1 Introduction
- 1.1 Related Work
- 1.2 Dataset
- 2 Predictive Model
- 2.1 Knowledge Representation
- 2.2 Neural Architecture
- 3 Experiments
- 3.1 Human-Level Performance
- 3.2 Discussion
- 4 Conclusion
- References
- Predicting the Receivers of Football Passes
- 1 Introduction
- 2 Data Exploration
- 3 Methodology for Predicting the Receivers of Football Passes
- 3.1 Feature Extraction
- 3.2 Modeling Approach
- 3.3 Baseline Models
- 3.4 Evaluation Approaches
- 3.5 Feature Importance
- 4 Results
- 4.1 RQ1: How Well Can We Model the Receiver of a Pass?
- 4.2 RQ2: What Are the Important Factors that Explain the Receiver of a Pass?
- 5 Conclusions
- References
- Author Index
System requirements
File format: PDF
Copy protection: Watermark-DRM (Digital Rights Management)
System requirements:
- Computer (Windows; MacOS X; Linux): Use the free software Adobe Reader, Adobe Digital Editions, or any other PDF viewer of your choice (see eBook Help).
- Tablet/Smartphone (Android; iOS): Install the free app Adobe Digital Editions or another reading app for eBooks, e.g., PocketBook (see eBook Help).
- E-reader: Bookeen, Kobo, Pocketbook, Sony, Tolino and many more (only limited: Kindle).
The file format PDF always displays a book page identically on any hardware. This makes PDF suitable for complex layouts such as those used in textbooks and reference books (images, tables, columns, footnotes). Unfortunately, on the small screens of e-readers or smartphones, PDFs are rather annoying, requiring too much scrolling.
This eBook uses Watermark-DRM, a „soft” copy protection. This means that there are no technical restrictions to prevent illegal distribution. However, there is a personalised watermark embedded in the eBook that can be used to identify the purchaser of the eBook in the event of misuse and to provide evidence for legal purposes.
For more information, see our eBook Help page.