Football Match Prediction using Exploratory Data Analysis & Multi-Output Regression

Football is one of the world's most popular and highly spectated games. The unpredictability of a football match is what makes this sport special and loved. Crowd influences, home team advantage, hostile away game atmosphere, the underdog wins, and comebacks make it such a hard game to predict....

Full description

Saved in:

Bibliographic Details
Published in:	2022 IEEE Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI) pp. 1 - 6
Main Authors:	Majumdar, Ayush, Kaur, Ravneet, Kulkarni, Tarak, Jiruwala, Mufaddal, Shah, Sneh, Pise, Nitin
Format:	Conference Proceeding
Language:	English
Published:	IEEE 21-12-2022
Subjects:	Boosting Data analysis Exploratory Data Analysis Games Machine Learning Machine learning algorithms Multi-Output Regression Prediction algorithms Score Prediction Technological innovation Training
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Football is one of the world's most popular and highly spectated games. The unpredictability of a football match is what makes this sport special and loved. Crowd influences, home team advantage, hostile away game atmosphere, the underdog wins, and comebacks make it such a hard game to predict. This paper represents a detailed study of predicting the outcome of a football match and thus, in turn, predicting the winners of the upcoming 2022 FIFA World Cup using Exploratory Data Analysis (EDA) to determine the important features of the dataset and then training various machine learning techniques to predict the score. The algorithms tested are Random Forests, Decision Trees, K-Nearest Neighbors, XGBoost, and Gradient Boosting. In the context of international football matches, the findings of this comparative analysis revealed that the XGBoost and Gradient Booster produced the highest average accuracy of 98.34 per cent.
DOI:	10.1109/IATMSI56455.2022.10119340