THE EFFECTIVENESS OF DIFFERENT MACHINE LEARNING ALGORITHMS ON BASKETBALL PLAYERS’ SHOOTING PERFORMANCE

Year 2019, Volume: 10 Issue: 3, 256 - 269, 16.12.2019

Serpil Kılıç Depren

https://doi.org/10.17155/omuspd.507797

Cited By: 1

Abstract

Bu çalışmanın
temel amacı, National Basketball Association (NBA) oyuncularının atış isabeti
üzerinde hangi faktörlerin önemli bir rolü olduğunu belirlemektir. Bu amaca
ulaşmak için, çalışmada 2014-2015 NBA sezonunda oynanan her bir maç için oyuncu
bazlı ham veri seti kullanılmıştır. Yedi farklı makine öğrenme algoritması
uygulanmış ve aynı zamanda aşırı uyum problemini önlemek için 10 kat çapraz
geçerlilik prosedürü 10 defa tekrar edilmiştir. Analizde dokuz adet bağımsız
değişken ve bir ikili bağımlı değişken kullanılmıştır. Bir basketbol
oyuncusunun başarılı bir atış yapıp yapamayacağını tahmin etmek için kullanılan
algoritmalar arasında en başarılı makine öğrenme algoritması k-en yakın komşu
algoritmasıdır. Atış Mesafesi, en yakın savunma oyuncusunun mesafesi ve temas
süresi oyuncunun başarılı bir atış yapmasını etkileyen en önemli faktörler
olarak tanımlanır. Oyuncuların atış performansı oyunu kazanmada çok etkili
olduğu için, bu çalışmanın sonuçları basketbol oyuncularına ve takım koçlarına
antrenman programları için bir rehber olarak kullanılabilir.

Keywords

Basketbol oyunu, makine öğrenmesi, performans analizi, sınıflama teknikleri

THE EFFECTIVENESS OF DIFFERENT MACHINE LEARNING ALGORITHMS ON BASKETBALL PLAYERS’ SHOOTING PERFORMANCE

Abstract

The main
purpose of this study is to determine which factors have an important role in National
Basketball Association (NBA) players’ shooting accuracy. To achieve this purpose,
player-based raw-dataset for each match
on the 2014-2015 NBA season is used in
this study. Seven different machine learning algorithms are applied and also
10-fold cross-validation with 10-repeat process is performed to avoid the overfitting problem. Nine independent variables
and one binary dependent variable are included in the analysis. According to
the results of the analysis, k-nearest neighbor
algorithm is the best machine learning algorithm among other algorithms that
are used in the analysis in order to predict whether basketball player can make
a shot or not. Shot Distance, distance of
closest defense player and touch time are
identified as the most important factors affecting player’s successful field
goal accuracy. Since the successful field goal
performance is very influential in winning the game, the results of this study
can be used as a guide for training programs to basketball players and team coaches. 

Keywords

Basketball games, classification techniques, machine learning algorithms, performance analysis

References

• 1. Hughes M, and Franks IM. Notational analysis of sport systems for better coaching and performance in sport. London: Routledge, 2004.
• 2. Leite N, Baker J, and Sampaio J. Paths to expertise in Portuguese national team athletes. Journal of Sports Science and Medicine, 2009; 8(4): 560-566.
• 3. Ortega E, Villarejo D, and Palao J. Differences in game statistics between winning and losing rugby teams in the six nations tournament. Journal of Sports Science and Medicine, 2009; 8(4): 523-527.
• 4. Bartlett R. Performance analysis: can bringing together biomechanics and notational analysis benefit coaches? International Journal of Performance Analysis in Sport, 2001; 1(1): 122-126.
• 5. Hughes M, and Bartlett R. The use of performance indicators in performance analysis. Journal of Sports Sciences, 2002; 20: 739-754.
• 6. Trninic S, Dizdar D and Luksic E. Differences between winning and defeated top quality basketball teams in final of European club championship. Collegium Antropologicum, 2002; 26(2): 521-531.
• 7. Hughes M, and Franks IM. The essentials of performance analysis – An introduction. London: Routledge, 2008.
• 8. Tsamourtzis E, Karypidis A, and Athanasiou N. Analysis of fast breaks in basketball. International Journal of Performance Analysis in Sport, 2005; 5(2): 17-22.
• 9. Csataljay G, O’Donoghue P, Hughes M, et al. Performance indicators that distinguish winning and losing teams in basketball. International Journal of Performance Analysis in Sport, 2009; 9(1): 60-66.
• 10. Zuccolotto P, Manisera M, and Sandri M. Big data analytics for modeling scoring probability in basketball: The effect of shooting under high-pressure conditions. International Journal of Sports Science and Coaching, 2017; 13(4): 569-589.
• 11. Sampaio J, Janeira M, Ibáñez S, et al. Discriminant analysis of game-related statistics between basketball guards, forwards and centres in three professional leagues. European Journal of Sport Science, 2006; 6(3): 173-178.
• 12. Ibáñez SJ, Sampaio J, Feu S, et al. Basketball game-related statistics that discriminate between teams’ season-long success. European Journal of Sport Science, 2008; 8(6): 369-372.
• 13. Puente C, Coso JD, Salinero JJ, et al. Basketball performance indicators during the ACB regular season from 2003 to 2013. International Journal of Performance Analysis in Sport, 2015; 15(3): 935-948.
• 14. Casals M, and Martinez AJ. Modelling player performance in basketball through mixed models. International Journal of Performance Analysis in Sport, 2013; 13(1): 64-82.
• 15. Gryko K, Mikołajec K, Maszczyk A, et al. Structural analysis of shooting performance in elite basketball players during FIBA EuroBasket 2015. International Journal of Performance Analysis in Sport, 2018; 18(2): 380-392.
• 16. Hosmer D, and Lemeshow S. Applied Logistic Regression (2nd ed.). Hoboken, NJ: John Wiley & Sons, Inc., 2000.
• 17. Yang CC, Soh CS and Yap VV. A non-intrusive appliance load monitoring for efficient energy consumption based on Naive Bayes classifier. Sustainable Computing-Informatics and Systems, 2017; 14: 34-42.
• 18. Kılıç Depren S. Prediction of Students’ Science Achievement: An Application of Multivariate Adaptive Regression Splines and Regression Trees. Journal of Baltic Science Education, 2018; 17(5): 887-903.
• 19. Nieto PG, Garcia-Gonzalo E, Anton JA, et al. A comparison of several machine learning techniques for the centerline segregation prediction in continuous cast steel slabs and evaluation of its performance. Journal of Computational and Applied Mathematics, 2017; 330(1): 1-19.
• 20. Ayyıldız E, Purutçuoğlu V, and Weber GW. Loop-based conic multivariate adaptive regression splines is a novel method for advanced construction of complex biological networks. European Journal of Operational Research, 2018; 270(3): 852-861.
• 21. Jiang S, Pang G, Wu M, et al. An improved K-nearest-neighbor algorithm for text categorization. Expert Systems with Applications, 2012; 39(1): 1503-1509.
• 22. Liu S, and Meng L. Re‐examining factor structure of the attitudinal items from TIMSS 2003 in cross‐cultural study of mathematics self‐concept. Educational Psychology, 2010; 30(6): 699-712.
• 23. Khun M, and Johnson K. Applied Predictive Modeling. New York: Springer, 2013.
• 24. Han J, Kamber M, and Pei J. Data Mining Concepts and Techniques. Waltham: USA: Elsevier Inc., 2012.