I am interested in learning of any studies using human subjects concerning
how humans reason with existing domain knowledge to learn something about a
particular case. For example, I would not be interested in a study
concerning learning causes from data. However, I would be interested in a
study in which the human already had learned some causal structure and
reasons with that structure. For example, the human might know smoking
causes both lung cancer and yellow teeth, and that tuberculosis and lung
cancer both cause dyspnea. We might test whether the human realizes that,
to learn whether Joe has lung cancer, it is informative to learn whether
his teeth are yellow but it is not informative to learn whether he has
tuberculosis.
I am interested not only in studies involving reasoning with causes, but in
any studies which concern reasoning with existing domain knowledge to learn
something about a particular case.
Thanks,
Rich
Richard E. Neapolitan
Professor and Chair
Computer Science Dept.
Northeastern Illinois University
5500 N. St. Louis
Chicago, Il 60625
http://www.neiu.edu/~reneapol/renpag1.htm
_______________________________________________
uai mailing list
[email protected]
https://secure.engr.oregonstate.edu/mailman/listinfo/uai
