/Basketball-Playoff-Prediction

A program to predict playoff participation in NBA

Primary LanguagePython

Basketball-Playoff-Prediction

Aim of this program

The program uses the NBA dataset from Kaggle (https://www.kaggle.com/fivethirtyeight/fivethirtyeight-nba-elo-dataset?select=nbaallelo.csv), asks for a team's first 10 results and predicts whether the team participate in the playoff or not.

Structure and operation

The program contains three different files (modules).
The first one is responsible for feature engineering and data preparation for machine learning algorithms.
The second file has two classes: train and test. These classes applies different classification algorithms (Decision Trees, Random Forests, Support Vector Classifiers) for training and testing the dataset.
The third file creates a user menu and asks for input - these are the results of the first 10 matches of the season.

The program gives back whether the given team would reach the playoff or not.

Technology and Methodology


For creating and preparing the data: Python Pandas
For training and testing: Python Scikit-Learn
The whole program has been written per OOP standards.