Analytics and machine learning approaches to generate insights for different sports

Abstract

Machine Learning and Data Analytics are used in many sectors so that it can help them to improve their services and find out the future predictions as well by using the previous data. One such sector that has been increasingly using this technology is sports. Many machine learning algorithms are available for sports prediction so that one can determine the team’s strength, weakness and predict the future outcome of the game. But these predictions are not always accurate. So the objective of this project is to implement common machine learning and analytics approach so that it can be used to predict the future outcomes of different games such as football, basketball and also generate insights for the same. Instead of using only one algorithm on the dataset of the scores from the previous matches, a series of an algorithm will be applied so that it can compare the final result from each algorithm and provide us with the most accurate result. Algorithms used will be the SVM model, NNR model, Random Forest Algorithm and ANN model. By generating the insights it is possible to not only determine the winner but also the position of the individual players on the field based on their respective performances. This project will thus predict the outcome of the games to a great extent which will help the teams to improve and turn their weaknesses into strength.