TY - GEN
T1 - Prediction Learning Model for Soccer Matches Outcomes
AU - Hervert-Escobar, Laura
AU - Matis, Timothy I.
AU - Hernandez-Gress, Neil
N1 - Funding Information:
∗The authors are grateful to the financial support of CONACYT-SNI program in order to promote quality research.
Publisher Copyright:
© 2018 IEEE.
PY - 2018/10
Y1 - 2018/10
N2 - Sport is a global business into which passionate fans and smart corporations pour hundreds of billions of dollars. Particularly, football soccer detonates a great movement of money in bets, sponsorships, attendance to parties, sale of t-shirts and accessories, etc. Professional soccer has been in the market for quite some time. The sports management of soccer is awash with data, which has allowed the generation of several metrics associated with the individual and team performance. The aim is to find mechanisms to obtain competitive advantages. In this paper, we propose a procedure for predicting the outcome of a soccer match. The procedure consist of a Bayesian Model based on rank position along with a Machine Learning Model based on historical data of matches. The procedure was tested using a data set containing the results of over 200,000 soccer matches from different soccer leagues around the world and for predicting the outcome of the FIFA world cup 2018. The results shoed an improvement in accuracy and rank probability error compared with other methodologies.
AB - Sport is a global business into which passionate fans and smart corporations pour hundreds of billions of dollars. Particularly, football soccer detonates a great movement of money in bets, sponsorships, attendance to parties, sale of t-shirts and accessories, etc. Professional soccer has been in the market for quite some time. The sports management of soccer is awash with data, which has allowed the generation of several metrics associated with the individual and team performance. The aim is to find mechanisms to obtain competitive advantages. In this paper, we propose a procedure for predicting the outcome of a soccer match. The procedure consist of a Bayesian Model based on rank position along with a Machine Learning Model based on historical data of matches. The procedure was tested using a data set containing the results of over 200,000 soccer matches from different soccer leagues around the world and for predicting the outcome of the FIFA world cup 2018. The results shoed an improvement in accuracy and rank probability error compared with other methodologies.
KW - Bayesian models
KW - Machine Learning
KW - Soccer
KW - prediction
KW - soccer matches
UR - http://www.scopus.com/inward/record.url?scp=85092068838&partnerID=8YFLogxK
U2 - 10.1109/MICAI46078.2018.00018
DO - 10.1109/MICAI46078.2018.00018
M3 - Conference contribution
AN - SCOPUS:85092068838
T3 - Proceedings of the Special Session - 2018 17th Mexican International Conference on Artificial Intelligence, MICAI 2018
SP - 63
EP - 69
BT - Proceedings of the Special Session - 2018 17th Mexican International Conference on Artificial Intelligence, MICAI 2018
A2 - Batyrshin, Ildar
A2 - de Lourdes Martinez Villasenor, Maria
A2 - Espinosa, Hiram Eredin Ponce
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 22 October 2018 through 27 October 2018
ER -