Work in Progress
This code implements a naive Bayesian classifier to help you find interesting stories from an RSS feed. (Currently, the Hacker News 20-point stories feed is hardcoded.)
To train the classifier, run
python train.py
After training, you can get predictions on most/least interesting stories by running
python predict.py
My data size is small, so it's not possible to see how well this works yet. You probably need to train it a lot.