Assessment of methods for predicting the NBA regular season MVP using Regression analysis and Classification

Predicting the NBA regular season MVP using regression and classification. An assessment of methods

The purpose of this project is to investigate different approaches within the Data Science discipline, that can be used to predict the NBA regular season MVP. The approaches that were explored were by using regression analysis and by converting the regression prediction problem into a binary classification problem. The algorithms used in both cases achieved acceptable results. Finding the best parameters for each model to help achieve the best possible predictive results is something beyond the scope of this project and therefore it was not explored in detail. This should be taken under consideration when evaluating the performance of the models at test.

The stacking ensemble of regression algorithms achieved the best performance amongst the models tested based on regression analysis.
For the classification approach the Support Vector Machine classifier achieved the best results.

Inside this repo you will find code artifacts and a report with its' corresponding presentation

Enjoy!