Please use this identifier to cite or link to this item:
Title: Football mobile application with built-in result prediction function using neural networks
Authors: Wang, Jia
Keywords: Engineering::Computer science and engineering
Issue Date: 2020
Publisher: Nanyang Technological University
Project: PSCSE18-0071
Abstract: Football is undoubtedly one of the most popular sports that connecting diverse communities, breaking down barriers of cultures, motived various levels of people. Football has likewise become one of the most profitable industries with the transfer market, broadcast authorization, sponsorship, and betting. There is a rising demand for a mobile application that allows fans to watch the latest updates of their favorite team and league. Moreover, numerous people stay interested in the prediction of the football match result. With the advantage of the latest machine learning technology, we can transform match result prediction as a multi-class classification problem with three class labels: win, draw, and lose with a perspective of the home team. This project aimed to develop a comprehensive, customizable mobile application that provides flexibility to users to pick their preferred team and league. In order to receive the latest news, fixtures, statistics, and standing status. It further contains a built-in prediction function for football match results based on the most recent line-ups announced by the two teams. This project will contain a Feedforward Neural Network model build through Python (Keras). With the help of historical data of each team obtained from an open-source database published at Kaggle, we can train the model and fit it to different input teams and their corresponding line-up players. The application displayed prediction made by this model. This report will discuss the process of developing the application as well as the methodology of finding the best varies of the model that can provide the prediction with the highest accuracy.
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
  Restricted Access
3.8 MBAdobe PDFView/Open

Page view(s)

Updated on Feb 5, 2023


Updated on Feb 5, 2023

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.