We present a case-based reasoning technique based on conceptual neighborhoods of cases. The system applies domain knowledge to the case retrieval process in the form of recall and generalize methods. Recall methods utilize domain specic preconditions and perform exact matching, while generalize methods apply transformations that generalize features in queries. The system uses a similarity function based on edit distances, where an edit distance considers only a subset of the features. This retrieval strategy enables the system to locate conceptually similar cases within the feature space. We demonstrate the performance of this approach by applying it to build-order selection in a real-time strategy game. Our results show that the system outperforms nearest neighbor retrieval when enforcing imperfect information in a real-time strategy game.
Conceptual Neighborhoods in Case-Based Reasoning
Conceptual Neighborhoods in Case-Based Reasoning
A paper by Ben G. Weber and Michael Mateas (University of California).
- Attachments
-
- iccbr_2009.pdf
- (159.38 KiB) Downloaded 456 times
Last edited by jarod42 on Sat Dec 29, 2012 12:53 am, edited 1 time in total.
Reason: add missing balise /quote
Reason: add missing balise /quote
04:27 jim4 why haven't you added wc1 support? this project sucks. i'm only going to use freecraft
05:06 jim4 finished wc1 support yet? i've been waiting for 6 years
05:10 jim4 new things scare me
05:06 jim4 finished wc1 support yet? i've been waiting for 6 years
05:10 jim4 new things scare me