This turns on the definition of "datamining." Surely scanning ostensibly private messages to update a database, even one built for innocuous purposes, especially without clear and conspicuous notice to users, is one form of it.

From the H-Online.com article: A spokesman for the company confirmed that it scans messages to filter out spam and phishing websites. This explanation does not appear to fit the facts, however. Spam and phishing sites are not usually found on HTTPS pages. By contrast, Skype leaves the more commonly affected HTTP URLs, containing no information on ownership, untouched. Skype also sends head requests which merely fetches administrative information relating to the server. To check a site for spam or phishing, Skype would need to examine its content.

I remember skype being known for making it possible to spy on user conversations, text and audio/video, way before microsoft bought it.

With microsoft moving skype back from p2p to centralized server architecture doesn't hints at making spying on users more difficult.

That's a separate issue. Obviously Skype could have MiTM connections since forever. They probably were legally required to do so, too. But decoding messages isn't the same as "datamining" them. And it's disingenuous, in context of IMs being datamined for advertising, to call an antispam service datamining, even if it's possibly technically true.