This is the Python implementation of the algorithm described in the paper included.
The code itself is found in the algorithm.py file.
Because I couldn't find a news API that collects data from the sources I'd analysed and had both headlines and content enough of the time, so I chose to run the algorithm on a sample set I wrote. And in recognition of my laziness, I forwent actually writing content, and just skipped to the keywords themselves. The code for extracting keywords does work, however, if you want to try it out.