[jskherman]

There's also libraries like trafilatura in Python featured here in HN some time ago that could extract content from websites to help augment the data.