The saga continues from where I left off(https://www.martineve.com/2017/01/16/thinking-more-about-eu-law-and-uk-copyright-exemptions/). Since then, I emailed a publisher to request a corpus of a specific author's work in a format that would allow computational techniques (i.e. not Amazon Kindle, which has DRM protection that it is illegal to break). Sadly, the publisher refused on the grounds of complexity, so I am no closer to being able to
Martin Paul Eve (Tue,) studied this question.