In this paper, we used a research approach based on grounded theory in order to classify methods proposed in literature that try to extend the verification of classical rule bases to the case of fuzzy knowledge modeling. Within this area of verification we identify two dual lines of thought respectively leading to what is termed respectively static and dynamic anomaly detection methods. The major outcome of the confrontation of both approaches is that their results, most often stated in terms of necessary and/or sufficient conditions are difficult to reconcile. This paper addresses precisely this issue by the construction of a theoretical framework, which enables to effectively evaluate the results of both static and dynamic verification theories. Things essentially go wrong when in the quest for a good affinity, matching or similarity measure, one neglects to take into account the effect of the implication operator, an issue that rises above and beyond the fuzzy setting that initiated the research. The findings can easily be generalized to verification issues in any knowledge coding setting.
展开▼
