NicolasAG / MachineLearning-project4

COMP 598 - Applied Machine Learning - Final project: Machine Learning for Improving Healthcare

The goal of this project is to investigate healthcare data. We began by identifying an interesting prediction question that can be tackled using machine learning methods. The problem decided was: predict the motor and total UPDRS scores for people with early-stage Parkinson's disease (cf: https://archive.ics.uci.edu/ml/datasets/Parkinsons+Telemonitoring) We then solved the problem using appropriate machine learning algorithms and methodology.

Requirements:

Python libraries:

datetime
random
csv
numpy
sklearn
...

To run the algorithms:

• Linear & Polynomial Regression:

  • File structure requirements:

  	data.csv
      NicoRegression.py
  

  • Check NicoRegression.py for parameter selection, and when ready run the following command: python NicoRegression.py

• SVM:

  • File structure requirements:

      data.csv
      svm.py
  

  • Check svm.py for parameter selection, and when ready run the following command: python svm.py

• Neural Network:

  • File structure requirements:

      data.csv
      NN_ML.py
  

  • Check NN_ML.py for parameter selection, and when ready run the following command: python NN_ML.py

Authors:

• Andres Felipe Rincón
• Ryan Razani
• Nicolas Angelard-Gontier