No abstract

Case-based reasoning (CBR) means reasoning from prior examples and it has considerable potential for building intelligent assistant system for the World Wide Web. In order to develop successful Web-based CBR systems, we need to select a set of representative cases for the client side case-base such that this thin client is competence in problem solving. This paper proposes a fuzzyrough method of selecting cases for such a distributed CBR system, i.e., a thin client system (a smaller case-base with rules) connected to a comparatively more powerful server system (the entire original case-base). The methodology is mainly based on the idea that an original casebase can be transformed into a smaller case-base together with a group of fuzzy adaptation rules, which could be generated using our fuzzy-rough approach. As a result, the smaller case-base with a group of fuzzy rules will almost have the same problem coverage as the entire original case-base. The method proposed in this paper, consists of four steps. First of all, an approach of learning feature weights automatically is used to evaluate the importance of different features in a given case-base. Secondly, clustering of cases is carried out to identify different concepts in the case-base using the acquired feature weights. Thirdly, fuzzy adaptation rules are mined for each concept using a fuzzy-rough method. Finally, a selection strategy which based on the concepts of case coverage and reachability is used to select representative cases. The effectiveness of our method is demonstrated experimentally using some testing data in the travel domain.Keywords Case-base maintenance, Fuzzy set, Rough set, Distributed case-based reasoning IntroductionRecently, with the rapid growth of applying intelligent systems on the World-Wide Web (WWW), practitioners are becoming more interested to consider the potential of distributed Case-based reasoning (CBR) systems. For example, consider a distributed CBR help-desk application over the WWW in which the representative cases need to be physically stored in the client nodes (i.e. for quick access), and all the original cases are resided in the central server. In order to keep this distributed CBR system upto-date, we need a strategy to maintain the cases as well as the reasoning technique (i.e. similarity function) used. As pointed out by Smyth [1], it is always better to have a local copy of a portion of the case-base (i.e. representative cases) stored on the client side. This would improve system performance and robustness as well as reduce substantially the load on the server and the network.For real world problems, there is always a limit on the size of the client side case-bases, a limit that is based on space and performance trade-off. The question arises then is: how to select cases for storing on the client side? Usually, this client side case-base will act as a cache, i.e. storing cases that are most likely to be needed by the individual users. A simple solution is to store those cases that are similar to the kind of probl...

No abstract

No abstract