Training Artificial Neural Networks – League Case

Artificial neural networks applied: Imagine a match between our team and 4 more. Our team have played already 3 rounds with each one, winning (0) or losing (1).

The match historic is like this:

12 4 1

1 0 0 0 1
0 1 0 0 0
0 0 1 0 0
0 0 0 1 1

1 0 0 0 0
0 1 0 0 1
0 0 1 0 0
0 0 0 1 1

1 0 0 0 0
0 1 0 0 0
0 0 1 0 1
0 0 0 1 1

Each row is when our team plays with other. The first 4 columns indicated which what team is competing.

The fifth column indicates if our team won (0) or lost (1). So after training the network, we can attempt to forecast the next matches:

If our team plays against the first team (1 0 0 0), it has a chance of 0.34344059228897095. As it is closer to 0 than 1, our team has chances to win them.

If our team plays against the forth team (0 0 0 1), it has a chance of 1. Basically our network is telling us that our team is going to lose.

 

The trained network can be found in https://fann.ehehdada.com/ouBmGl4EvR4zntnIW

 

Advertisements
%d bloggers like this: