INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
An ANN Model for Prediction of Goal Scores by Individual Team
Authors Name:
Omega sarjiyus
, Benson Yusuf Baha , Eric Jones
Unique Id:
IJSDR2308087
Published In:
Volume 8 Issue 8, August-2023
Abstract:
Football is the most popular sport in the world, played in most countries in the world. The sport has a very large betting industry. The current estimations, which include both the illegal markets and the legal markets, suggest the sports match-betting industry is multi-billion dollar industry. This means the ability to accurately predict outcomes of games can be very lucrative. Although various research works have employed statistical and mathematical models to predict match results, none of the models are able to predict the exact number of goals to be scored by each team in a football match. In this research, an Artificial Neural Network (ANN) model was developed following the Machine Learning Life Cycle. The dataset was obtained from Football-Data.co.uk, a reputable football data website. Backward Elimination technique was used to select the dataset features for the proposed neural network model. The features selected include: h_a= Home or Away, xG= expected goals, xGA = expected goals away team, npxG = expected goals without penalties, npxGA = expected goals without penalties away team, deep = Number of plays in opponent final third, scored = goals scored, missed = goals conceded, result = win, draw or lose date = Date of the match, wins = binary for wins, draws = binary for draws, loses = binary for loses, teamId = team name, matchtime = time of match, tot_goal = total goals team has scored so far, tot_con = total number goals conceded by the team so far, Referee.x = refree name, HtrgPerc = shot on target/total shots – Home, AtrgPerc = shot on target/total shots – Away, matchDay = day of match. The dataset was then split into training and testing set at 75% and 25% respectively. The neural network developed is a multi-layer perceptron classifier implemented by the MLPClassifier class in sklearn. The model was compiled with different parameters to find the model with the highest accuracy relative to the mean squared error. The graph for the accuracy score and mean squared error was plotted and it showed the mean squared error was relatively the same for all the models. The model with the highest accuracy score was selected. The selected model has three (3) hidden layers that consist of 10,10, and 10 neurons with sigmoid optimizer and tanh activation function.. The model ran 1000 epochs and got an accuracy score of 97.92% with MSE of 2.8644, implying that real life games with unknown results can indeed be predicted with a high level of accuracy using machine learning.
"An ANN Model for Prediction of Goal Scores by Individual Team", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.8, Issue 8, page no.594 - 599, August-2023, Available :http://www.ijsdr.org/papers/IJSDR2308087.pdf
Downloads:
000338721
Publication Details:
Published Paper ID: IJSDR2308087
Registration ID:206992
Published In: Volume 8 Issue 8, August-2023
DOI (Digital Object Identifier):
Page No: 594 - 599
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn