Hey! Thanks for submitting to Towards Data Science. I've been over your article and I really like it - but I think it'd be helped by adding a bit of code to help the reader understand some steps, and also, a bit of rationale as to why you used a regression instead of say, a Random Forest?
You've done lots of great work here - it'd just be even better if you shared some code/formulaes on how you scraped the sites or how you ran your ML algos.
Otherwise, all looks good! Great job!