Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy...

13
Alberto Trindade Tavares Predicting Results of Brazilian Soccer League Matches ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems

Transcript of Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy...

Page 1: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

Alberto Trindade Tavares

Predicting Results of Brazilian Soccer League

Matches

ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems

Page 2: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

Since 2003, the Brazilian Soccer League has the following format: 20 participating clubs Each club faces every other club twice in the season, once at their

home stadium, and once at that of their opponents 380 matches divided into two parts:

First half: May-August Second half: September-December

A match has three possible results: Win of the home team Draw Loss of the home team

Brazilian Soccer League

Page 3: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

Predict the outcome (win of home team, draw, or loss of home team) of every game of the second half for the current season (2013)

Using as training data the game results of the first half of 2013 season

Develop two classifiers, using MATLAB, for performing these predictions : Maximum Likelihood Classifier

Multi-Layer Perceptron

Compare their accuracy between themselves and to other works

Goal of this Work

Page 4: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

The feature vector for representing a match instance has six features, the first three for the first team (home), and last three for the second team (visiting):

Results from 2003 season to the last match of current season

Feature Vector

# wins as home team

# draws as home team

# losses as home team

# wins as visiting team

# draws as visiting team

# losses as visiting team

First Team Second Team

Page 5: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

Extraction of results of every match since 2003

Two different sources: 2003-2004 seasons:

http://www.bolanaarea.com/gal_brasileirao.htm 2005-2013 seasons: http://www.campeoesdofutebol.com.br

Python program for parsing the HTML pages, and storing the results into text files, which can be read via MATLAB function load

Data Extraction

Page 6: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

Gaussian Distribution

Maximum Likelihood Classifier

x

P(x)Win

DrawLoss

Page 7: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

Classification rate per round:

Maximum Likelihood Classifier (Results)

Average Classification Rate = 53.1579%

Page 8: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

Total Confusion Matrix:

Maximum Likelihood Classifier (Results)

Predicted Wins

71 8 18

25 9 15

19 4 21

Predicted Draws Predicted Losses

Actual Wins

Actual Draws

Actual Losses

Page 9: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

# Hidden Layers = 3

# Neurons in First Hidden Layer = 3

# Neurons in First Hidden Layer = 20

# Neurons in First Hidden Layer = 3

Learning rate (α) = 0.1

Momentum = 0

Hidden layers use hyperbolic tangent activation function, and output layer uses sigmoid activation function

Multi-Layer Perceptron

Page 10: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

Classification rate per round (10 runs):

Multi-Layer Perceptron(Results)

Average Classification Rate = 55.7895%

Page 11: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

Total Confusion Matrix:

Multi-Layer Perceptron(Results)

Predicted Wins

78 7 12

27 11 11

23 4 17

Predicted Draws Predicted Losses

Actual Wins

Actual Draws

Actual Losses

Page 12: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

A. Joseph, N.E. Fenton, M. Neil. Predicting football results using Bayesian nets and other machine learning techniques (2006)

Published in the Journal Knowledge-Based Systems

Their results: Naïve BN: 47.86% kNN: 50.58% Expert BN: 59.21%

Comparison with other work

My Results: Maximum Likelihood:

53.1579% Multi-Layer Perceptron:

55.7895%

Page 13: Alberto Trindade Tavares ECE/CS/ME 539 - Introduction to Artificial Neural Network and Fuzzy Systems.

Questions?