This is the companion resource for the article mentioned above. It contains everything you need to conduct similar experiments as described in the paper or use it in your own teaching.
This includes my tool for extracting the text and images from a large set of PDF files (the coursework):
PDF Conversion ToolThe results are stored in JSON files to facilitate processing. I have also prepared a live demo that allows you to analyze images “out of the box”:
Live Analysis ToolThe analyses are very rudimentary. You may need to tailor the setup to fit your specific needs. Finally, I have provided the source code for the experiment which you can adapt for your own purpose.
Download Source Code