Prediction of Football Matches’ Results Using Neural Networks Roger Achkar 1(&) , Ibraheem Mansour 1 , Michel Owayjan 2 , and Karim Hitti 3 1 Department of Computer and Communications Engineering, AUST, Achrafieh, Lebanon rachkar@aust.edu.lb 2 Department of Mechatronics Engineering, AUST, Achrafieh, Lebanon mowayjan@aust.edu.lb 3 Department of Mathematics, Faculty of Sciences and Issam Fares Faculty of Technology, University of Balamand, El-koura, Lebanon Abstract. In this paper, learning with a teacher artificial neural network to predict the results of football matches is presented. This type of networks requires training via examples, and when the training is complete, the network can be tested to check the results of new examples. In this application, the training examples are the results of previous matches which the network will use to predict the results of new ones. Keywords: Neural network  Multilayer perceptron  Activation function Weights  Learning parameter 1 Introduction The idea behind artificial neural networks is to produce a copy of human brain cells and their interconnections in order to make a network that behaves like the brain itself: learn new things, identify and classify patterns, and be able to take decisions just like human beings. Some types of neural networks need a teacher to learn, however, other types can learn by themselves without human intervention which makes neural net- works applicable for various types of applications [1]. They are used to perform many tasks that include classification, clustering, recognition of patterns, etc. Neural net- works are deep learning technologies; they resemble the human brain in two ways: • A neural network gains knowledge through learning, whether it was supervised, unsupervised, or reinforced. • This knowledge is stored in synaptic weights that are the inter-neuron connections [2–4]. Several types of neural networks exist today. The most popular one is the Multi- layer Perceptron (MLP), which belongs to a general class of structures called feed- forward neural networks. It is the type of network that will be implemented in the prediction application presented in this paper. © Springer Nature Switzerland AG 2019 A. E. Hassanien et al. (Eds.): AISI 2018, AISC 845, pp. 169–178, 2019. https://doi.org/10.1007/978-3-319-99010-1_15