Football Result Prediction using Dixon Coles Algorithm
Download as PPTX, PDF1 like1,998 views
This paper explores the use of machine learning, specifically artificial neural networks, for predicting football match outcomes by analyzing various performance attributes and expert ratings. It identifies key factors influencing match results, evaluates different predictive methodologies, and highlights challenges faced in sports prediction. The results indicate a notable correlation between expert ratings and match outcomes, although achieving high accuracy in predictions remains difficult.
More Related Content
Similar to Football Result Prediction using Dixon Coles Algorithm (20)
Football Result Prediction using Dixon Coles Algorithm
- 1. FOOTBALL PREDICTOR USING MACHINE LEARNING AAKASH M SRIRAM V VISHNU M
- 2. INTRODUCTION Machine learning (ML) is one of the intelligent methodologies that have shown promising results in the domains of classification and prediction. One of the expanding areas necessitating good predictive accuracy is sport prediction, due to the large monetary amounts involved in betting. In addition, club managers and owners are striving for classification models so that they can understand and formulate strategies needed to win matches. These models are based on numerous factors involved in the games, such as the results of historical matches, player performance indicators, and opposition information. This paper provides a critical analysis of the literature in ML, focusing on the application of Artificial Neural Network (ANN) to sport results prediction. In doing so, we identify the learning methodologies utilised, data sources, appropriate means of model evaluation, and specific challenges of predicting sport results.
- 3. OBJECTIVE 1. To identify the most important attributes of player’s performance which determine their ratings given by experts. In this way we are finding out the latent knowledge which the experts use to assign ratings to players. 2. To find out which performance attributes of the players of the two competing teams affect the match outcome and to what extend the match outcome is characterised by the performance attributes of players. 3. Find out good ways in which individual player ratings can be aggregated resulting in a set of team ratings and investigate how closely these team ratings can determine the match outcome. This will indicate the correlation between expert ratings and match outcomes. Hence, it will show the influence of match outcomes on the expert ratings. 4. To investigate how well the expert ratings given to the players of a team in the past performances of the team predict the next match outcome.
- 4. LITERATURE SURVEY Ranking of Sports Teams via the AHP The hierarchical structure is practical because most organizations and decision processes are complex and often are organized in hierarchical form. Accuracy of using Eigen-vector is less compared to neural networks and ML techniques. Predicting Football Match Results with Logistic Regression It’s advantages are it’s very suitable to explain the relationship between output variable and input, and it can solve the problem which ordinary least squares regression cannot. There is so many unexpected result in 2015/2016 records and when incorporated in training data, the model built was twisted and does not produce the expected result. A Survey of Content- Aware Video Analysis for Sports We review the developments in sports video analysis, focusing on content-aware techniques that involve understanding and arranging the video content on the basis of intrinsic and semantic concepts. Many open problems remain because of the diversity of game structures among sports domains. Developing a unified framework that enables processing data from diverse sports is still challenging.
- 5. ADVANTAGES ● Find a good ratings aggregation method. ● Using this aggregation strategy we find the set of attributes which can best characterise the match performance. ● Characterise the match outcome using the attributes generated from aggregated player ratings. ● Predict the outcome of next match using the attributes created from the aggregated ratings of player performances over the past matches. ● Best result for characterising match outcome was an accuracy of 90% using the best 8 attributes only. ● This shows a positive correlation between match outcome and the ratings given by the soccer experts.
- 6. DISADVANTAGES ● The best accuracy we obtained was 53.39% which is better than a random guess. ● This low accuracy was partly due to the extreme difficulty of predicting a ‘Draw’ for a match. ● This low accuracy also means that the outcome of a match depends only the players’ performances in the current match and not the performances of the past matches.
- 7. PROPOSED MODEL To increase the efficiency and to improve the probability ranking by : ● Using Dixon-Coles model and time waiting. ● Include goal times data. ● Add ranked probability score. ● Add standard errors to parameter estimates. ● Include more wide data sets such as physicality of players, experience, etc.
- 8. REQUIREMENTS ● Data set of Barclays Premier League (BPL). ● R language Studio.
- 22. REFERENCES ● Ranking of Sports Teams via the AHP by ZILLA SINU ANY -STERNA Department of Industrial Engineering and Management, Ben-Gurion University of the Negev, Beer Sheva, Israel, July 2016. ● Sync-Rank: Robust Ranking, Constrained Ranking and Rank Aggregation via Eigenvector and SDP Synchronization by Mihai Cucuringu, Jan 2018. ● Generic temporal features of performance rankings in sports and games by José A Morales, Sergio Sánchez , Jorge Flores , Carlos Pineda , Carlos Gershenson, Germinal Cocho, Jerónimo Zizumbo , Rosalío F Rodríguez and Gerardo Iñiguez, Springer Journal 2016. ● Predicting the winner of NFL-games using Machine and Deep Learning by Pablo Bosch, Feb 2018. ● A Network-Driven Methodology for Sports Ranking and Prediction by Vincent Xia, Kavirath Jain, Akshay Krishna, and Christopher G. Brinton, Department of Operations Research and Financial Engineering, Princeton University ,Department of Electrical Engineering, Princeton University, 2018.
- 23. REFERENCES ● Predicting Football Match Results with Logistic Regression by Darwin Prasetio,2017. ● IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, VOL. 28, NO. 5, MAY 2018 A Survey of Content-Aware Video Analysis for Sports Huang-Chia Shih, Member, IEEE ● Developing Analytical Tools to Impact U.Va. Football Performance by Jack Corscadden, Ross Eastman, Reece Echelberger, Connor Hagan, Clark Kipp, Erik Magnusson, Graham Muller, Stephen Adams, James Valeiras, and William T. Scherer, 2018. ● Predicting sports results using latent features: a case study Stefan Dobravec , May 2018. ● https://www.transfermarkt.co.uk/
- 24. BELIEVE THAT.