Please use this identifier to cite or link to this item:
Author(s): Tiago Filipe Mendes Neves
Title: A data mining approach to predict probabilities of football matches
Issue Date: 2019-07-11
Abstract: With the increasing growth of the amount of money invested in sports betting markets it is important to verify how far the machine learning techniques can bring value to this area. A performance evaluation of the state-of-art algorithms is performed and evaluated according to several metrics, incorporated in the CRISP-DM methodology that goes from web-scraping through to generation and selection of features. It is also explored the universe of ensemble techniques in an attempt to improve the models from the point of view of bias-variance trade-off, with a special focus on neural network ensembles.
Description: Com um crescimento cada vez maior dos volumes apostados em competições desportivas torna-se importante verificar até onde as técnicas de aprendizagem computacional conseguem trazer valor a esta área. É feita uma avaliação da performance de algoritmos estado-da-arte em diversas métricas, incorporado na metodologia CRISP-DM que é percorrida desde a aquisição de dados via web-scraping, passando pela geração e seleção de features. É também explorado o universo de técnicas de ensemble numa tentativa de melhorar os modelos do ponto de vista do bias-variance trade-off, com especial foco nos ensembles de redes neuronais.
Subject: Engenharia electrotécnica, electrónica e informática
Electrical engineering, Electronic engineering, Information engineering
Scientific areas: Ciências da engenharia e tecnologias::Engenharia electrotécnica, electrónica e informática
Engineering and technology::Electrical engineering, Electronic engineering, Information engineering
TID identifier: 202398200
Document Type: Dissertação
Rights: openAccess
Appears in Collections:FEUP - Dissertação

Files in This Item:
File Description SizeFormat 
343145.pdfA data mining approach to predict probabilities of football matches5.68 MBAdobe PDFThumbnail

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.