A compound framework for sports results prediction: A football case study

Min, Byoung-Goo; Kim, Jinhyuck; Choe, Chongyoun; Eom, Hyeonsang; McKay, Bob

doi:10.1016/j.knosys.2008.03.016

Cited by 84 publications

(41 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…al., 2006) demonstrated the importance of supplementing data with expert judgement by showing that an expert constructed Bayesian network was more accurate in generating football match forecasts for matches involving Tottenham Hotspurt than machine learners of MC4, naive Bayes, Bayesian learning and K-nearest neighbour. A model that combined a Bayesian network along with a rule-based reasoner appeared to provide reasonable World Cup forecasts in (Min et al, 2008) through simulating various predifined strategies along with subjective information, whereas in (Baio & Blangiardo, 2010) a hierarchical Bayesian network model that did not incorporate subjective judgments appeared to be inferior in predicting football results when compared to standard Poisson distribution models.…”

Section: Introductionmentioning

confidence: 99%

pi-football: A Bayesian network model for forecasting Association Football match outcomes

Constantinou

Fenton

Neil

2012

Knowledge-Based Systems

117

View full text Add to dashboard Cite

A Bayesian network is a graphical probabilistic belief network that represents the conditional dependencies among uncertain variables, which can be both objective and subjective. We present a Bayesian network model for forecasting Association Football matches in which the subjective variables represent the factors that are important for prediction but which historical data fails to capture. The model (pi-football) was used to generate forecasts about the outcomes of the English Premier League (EPL) matches during season 2010/11 (but is easily extended to any football league). Forecasts were published online at www.pi-football.com prior to the start of each match. In this paper, we demonstrate that a) using an appropriate measure of forecast accuracy, the subjective information improved the model such that posterior forecasts were on par with bookmakers' performance; b) using a standard profitability measure with discrepancy levels at ≥ 5%, the model generates profit under maximum, mean, and common bookmakers' odds, even allowing for the bookmakers' built-in profit margin.Hence, compared with other published football forecast models, pi-football not only appears to be exceptionally accurate, but it can also be used to 'beat the bookies'.

show abstract

Section: Introductionmentioning

confidence: 99%

pi-football: A Bayesian network model for forecasting Association Football match outcomes

Constantinou

Fenton

Neil

2012

Knowledge-Based Systems

117

View full text Add to dashboard Cite

show abstract

“…The system has some similarities with [13], although the reasoning system based on FCA is qualitative while the cited system is hybrid (bayesian reasoning). Pure qualitative reasoning was selected based on the aim of discovering trends (under a contextual selection) represented in the form of association rules with high confidence.…”

Section: Concluding Remarks and Future Workmentioning

confidence: 99%

“…Usually, one can work with hybrid models, and rarely with pure qualitative and exogenous reasoning systems appear in literature, although their use is considered for experiments (for example, frugal methods [3] and based on the recognition heuristic [10]) or as part of hybrid systems (see e.g. [13]). There are two reasons that may justify this point.…”

Section: Introductionmentioning

confidence: 99%

Confidence-Based Reasoning with Local Temporal Formal Contexts

Aranda-Corral

Borrego-Díaz

Galán-Páez

2011

Advances in Computational Intelligence

View full text Add to dashboard Cite

Abstract. Formal Concept Analysis (FCA) is a theory whose goal is to discover and to extract Knowledge from qualitative data. It provides tools for reasoning with implication basis (and association rules). In this paper we analyse how to apply FCA reasoning to increase confidence in sports betting, by means of detecting temporal regularities from data. It is applied to build a Knowledge based system for confidence reasoning.

show abstract

“…Machine learning methods have been successfully applied in football. For example, in the prediction of match outcomes (Constantinou, Fenton, & Neil, 2012;Min, Kim, Choe, Eom, & McKay, 2008;Odachowski & Grekow, 2013;Strnad, Nerat, & Kohek, 2015;Tüfekci, 2016), analysis of team performance (Arruda Moura, Barreto Martins, & Augusto Cunha, 2013) or player's injury prediction (Arndt & Brefeld, 2016;Jelinek, Kelarev, Robinson, Stranieri, & Cornforth, 2014;Kampakis, 2011). However, the problem of characterizing and selecting players based on available data of performance using machine learning methods is an interesting and open field of research today.…”

Section: Introductionmentioning

confidence: 99%

A Gaussian mixture clustering model for characterizing football players using the EA Sports' FIFA video game system. [Modelo basado en agrupamiento de mixturas Gaussianas para caracterizar futbolistas utilizando el sistema de videojuegos FIFA de EA Sports].

Soto-Valero¹

2017

Rev. int. cienc. deporte

View full text Add to dashboard Cite

The generation and availability of football data has increased considerably last decades, mostly due to its popularity and also because of technological advances. Gaussian mixture clustering models represents a novel approach to exploring and analyzing performance data in sports. In this paper, we use principal components analysis in conjunction with a model-based Gaussian clustering method with the purpose of characterizing professional football players. Our model approach is tested using 40 attributes from EA Sports' FIFA video game series system, corresponding to 7705 European players. Clustering results reveal a clear distinction among different performance indicators, representing four different roles in the team. Players were labeled according to these roles and a gradient tree boosting model was used for ranking attributes regarding to its importance. We found that the dribbling skill is the most discriminating variable among the different clustered players' profiles.Key words: association football; EA Sports' FIFA video game series system; machine learning; principal component analysis; Gaussian mixture clustering models; classification and regression trees. AbstractCorrespondencia/correspondence: César Soto-Valero Department of Computer Science, Universidad Central "Marta Abreu" de Las Villas, Cuba Email: cesarsotovalero@gmail.com En las últimas décadas se ha visto un incremento considerable en la generación y disponibilidad de datos de fútbol, esto se debe fundamentalmente a la popularidad de este deporte así como a los avances tecnológicos realizados en este campo. Los modelos de agrupamiento basados en mixturas Gaussianas representan un enfoque novedoso para explorar y analizar datos de desempeño deportivo. En el presente trabajo, se lleva a cabo una caracterización de jugadores profesionales de fútbol utilizando técnicas de análisis de componentes principales y agrupamiento basados en mixturas Gaussianas. El modelo presentado es comprobado utilizando datos del sistema de videojuegos FIFA de EA Sports, dichos datos representan 40 atributos correspondientes a 7705 futbolistas europeos. Los resultados del agrupamiento revelan una clara distinción entre algunos indicadores de desempeño, los cuales corresponden a cuatro roles diferentes en el equipo. Consecuentemente, los jugadores fueron etiquetados de acuerdo a estos roles y un modelo de árboles de gradiente ampliado fue utilizado para ordenar los atributos de acuerdo a su importancia. Como resultado se identificó a la habilidad de driblear como la variable que mejor discrimina entre los diferentes perfiles de jugadores.Palabras clave: fútbol; sistema de videojuegos FIFA de EA Sports; aprendizaje automátic; análisis de componentes principales; agrupamiento basado en modelos de mixturas Gaussianas; árboles de clasificación y regresión. Resumen

show abstract

A compound framework for sports results prediction: A football case study

Cited by 84 publications

References 14 publications

pi-football: A Bayesian network model for forecasting Association Football match outcomes

pi-football: A Bayesian network model for forecasting Association Football match outcomes

Confidence-Based Reasoning with Local Temporal Formal Contexts

A Gaussian mixture clustering model for characterizing football players using the EA Sports' FIFA video game system. [Modelo basado en agrupamiento de mixturas Gaussianas para caracterizar futbolistas utilizando el sistema de videojuegos FIFA de EA Sports].

Contact Info

Product

Resources

About