The data-source are the enormous data breach that are more and more frequent. There is more intensive to collect more information on someone you already know something about than spamming an email you don't even know if it's a valid one.
The spam can also be very more effective as it present itself with personal information about the spammed.
Edit: that’s not to deny that big data leaks are a serious problem
It's obvious to any non native english speaker, when you have a spam in english, it is because they toke the email from the web. When it's in you native language, it's usually from a data breach.
I'm vastly more spammed by the later. I can confirm it with unique email addresses of the "+" form (but not with the + character).
Also when I'm spammed in english, it's for Web3 crypto stuff and from a data breach it's a phishing attempt.
But yeah, I’d say most junk mail is coming to (1) an address leaked from one Russian bank (!) I used, (2) the address listed in public business databases (I have a company in Estonia).
That solution doesn't apply to the use case in the article.