Forums before death by AOL, social media and spammers... "We can't have nice things"
|    comp.ai    |    Awaiting the gospel from Sarah Connor    |    1,954 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 862 of 1,954    |
|    Steven Bogaerts to cyberGEEK    |
|    Re: How to evaluate/experiment a prototy    |
|    09 Dec 05 00:01:38    |
      From: sbogaert@indiana.edu              cyberGEEK wrote:       > Hi,       >       > Im new to case-based reasoning. Currently finished building my       > prototype system, it is a cbr recommender system for network fault.       > Built using IUCBRF, a cbr development framework from Indiana       > University.       >       > Im trying to test and document some results of it, but i have big       > problems here. I dont know how to test a cbr system, what to test, what       > component to test, what testing method to use, etc etc..I also dont       > know what to write in the experiment paper.       >       > If anyone could give me some advice or suggestion, it would be       > appreciated :)       >              Hi Wira, IUCBRF does in fact have experimentation facilities described       in the documentation that I believe you will find useful. As for what       exactly your experiments will look like and what the resulting paper       will be like, that depends on what particular problem you are examining.        I would recommend browsing other CBR papers for examples of how       experiments were conducted and written up. I hope that helps!              Steven Bogaerts       Indiana University       http://www.cs.indiana.edu/~sbogaert/CBR/index.html              [ comp.ai is moderated. To submit, just post and be patient, or if ]       [ that fails mail your article to |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca