F. Hoppner and F. Klawonn and R. Kruse and T. Runkler. (1999) Fuzzy Cluster Analysis -- Methods foc Classification, Data Analysis and Image Recognition. John Wiley and Sons  Home/Search   Document Not in Database   Summary   Related Articles

....directly [23] from data. The main advantage of rule based fuzzy classifiers over a DT is the greater flexibility of the decision boundaries. DT based classifiers carry out rectangular partitioning of the input space, while fuzzy models are able to generate non axis parallel decision boundaries [24]. Because of this, fuzzy systems are more parsimonious than DTs. Therefore, fuzzy models initialized simply from DTs [21] 22] could be more complex than it is needed. This suggest that a simple transformation of a DT into a fuzzy model is not enough, this step should be followed by a series of ....

....about the membership degrees of the classes is available that can give some hints about the ambiguousness of the decision. An another advantage of fuzzy systems is that they can easily define non axis parallel decision boundaries, while DTs always approximate such systems in a step wise manner [24] as it is shown in Figure 2. As this figure suggests, for an accurate approximation of a non axis parallel class, lot of crisp decision rules are needed, while a fuzzy model with two rules R 1 : If x 1 is A 11 and x 2 is A 12 then Class = 1 (7) R 2 : If x 1 is A 21 and x 2 is A 22 then Class = ....

F. Hoppner and F. Klawonn and R. Kruse and T. Runkler. (1999) Fuzzy Cluster Analysis -- Methods foc Classification, Data Analysis and Image Recognition. John Wiley and Sons

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC