Professor Roger Schank has contributed a lot to the study of natural language and Cognitive Science. He enlightened the field of natural language with his conceptual dependency theory and invented the concept of "script". The well-known NLP program SAM was constructed according to his 'script" scheme. He is the founder and director of The Institute for the Learning Sciences in Northwestern University.
1.) How would you comment on current NLP programs today?
They are basically statistical; they may work well in the short run, but the theoritcal basis of NLP has been abandoned in favor of short term win.
2.) In your book "The Cognitive Computer" you discuss the idea of "conceptual dependency". Do you believe that a successful machine translator will need to utilize CD in order to provide good results?
Or something like it; you can translate without understanding but you can't do it well.
3.) In that same book, you discuss the levels of computer understanding -- ranging from "making sense" to "cognitive understanding" to "complete empathy". When do you believe that computers will reach both "cognitive understanding" and "complete empathy", if at all.
Well it wont be soon; no real money is likely to be spent on making this kind of AI happen; computeres wont have massive knowledge or be able to really learn what they'd need to know to become really thinking machines unless someone is ready to spend a great dela of money on making that happen.
4.) Do you believe that "emotions" are necessary for "cognitive understanding" and "complete empathy"?
Probably.
5.) Do you think the "domain problem" will ever be overcome in the near future?
No; too much knowledge is needed even for one domain.
Professor