Drupal and Machine Learning: A Case Study
In this session, I'd like to explain how I used Drupal and open source machine learing libraries to collect and classify large amount of political news articles into left-leaning and right-leaning, which is a chapter of my dissertation. In particular, I'd like to cover these topics:
- What are the challenges of using Drupal with open source machine learning libraries (such as Apache Mahout, scikit, R, etc), and why?
- How to use Amazon Mechanical Turk to collect labeld data for classifier training?
- How to integrate non-PHP machine learning libraries into a Drupal site?
- What are some of the Drupal modules for machine learning purposes, and what are the cool things you might be able to do with them?
- A brief primer on machine learning algorithms.
The goal of the session is explore what you can do using machine learning libraries in Drupal, and how you do it. It's not about how to use any particular machine learning libraries. I will try to avoid using many jargons in machine learning.