3 ms·
This is great for spam!
by rw2 11y ago
This is great for spam!
- zbyte64 11y agoI know you meant "for producing spam" but it seems these techniques can also do semantic-relatedness[1] which I imagine could answer questions like: "Is this about viagra?" or "Is this related to something the mailbox owner actually talks about?" [1] https://github.com/ryankiros/skip-thoughts#semantic-relatedness https://github.com/ryankiros/skip-thoughts#semantic-relatedn...