Subject: This program uses supervised learning to anticipate students failures
How it works: A data set on students passing or failing the tests is available. The data is binarized and split into a training and testing set. A logistic regression is then applied to the training data. Note that a SVM or Gradient Boosting Trees could be used be the training and testing time is much higher. Finally, the model's testing score is determined with the testing set.
Running the program: lauch a console and type in the main directory: python student_intervention.py