"... In PAGE 4: ...b, because they create the rules completely out of training data. In Table1 the NEFCON model is compared to hybrid approaches that are only able to learn membership functions. The NEFCON architecture can be interpreted as a neural network and as a fuzzy controller.... In PAGE 4: ... For this reason it is suited for all kinds situations, where any kind of neuro{fuzzy model is applicable. Table1 displays under which circumstances the discussed generic models can be applied. If the decision for... ..."

"... In PAGE 2: ...ig. 1. Voltage peaks within 10 min. periods (maximal values). Experiment results are shown in Table2 . The neuro-fuzzy system was trained with 2000 training vectors consisting of 5 random values describing the disturbances within a 10 min.... In PAGE 2: ... The lower the output value the smaller the probability of a malfunction. Test input vectors and respective outputs are summarized in Table2 . In case 2 swells were severe, so the system output is close to 1.... ..."

"... In PAGE 7: ... George and Cardullo (1999) use NFS for prediction tasks within a simulated aircraft control. Table5 summarizes studies and applications of NFS for UM. NFS have been designed to retain the positive aspects of NN and FL, nevertheless it still maintains some of the limitations of both approaches, mainly the training time needed and application for dynamic modeling.... ..."

"... In PAGE 6: ... The neuro-fuzzy models are built according to the described framework and, to validate the proposed feature selection technique, a second group of neuro-fuzzy models has been defined with all 32 features. In Table1 the obtained results are listed together with a comparison of the two groups in terms of Mean Squared Error evaluated both on training and test set. The table reports also the name of the 22 output variables (chemical elements) we want to predict.... ..."

"... In PAGE 22: ...upports independent rules (i.e., changes in one rule do not effect the result of other rules). FSs and NNs differ mainly on the way they map inputs to outputs, the way they store information or make inference steps. Table5 lists the most popular software and hardware tools based on FSs as well as on merged FSs and NNs methodologies. Neuro-Fuzzy Systems (NFS) form a special category of systems that emerged from the integration of Fuzzy Systems and Neural Networks [65].... ..."

"... In PAGE 44: ... Table1 . Average classification success and standard deviation for the three models.... In PAGE 44: ... All differences found to be statistically significant with t values greater than 10. As we can see in Table1 , the FNN shows a better performance than the neuro-fuzzy model in the test set 1 (clear-cut cases) and the test set 2 (marginal cases). This performance of the neuro-fuzzy model is compensated from its performance in the test data 3 (special marginal cases).... In PAGE 52: ... Percentage of type of errors for the three models. Table1 . Average classification success and standard deviation for the three models.... ..."

"... In PAGE 20: ... ANFIS (Adaptive Neuro Fuzzy Inference System) was used to develop the neuro- fuzzy model (Jang, 1993). Table5 . Criteria values for the tool wear data (jackknife approach) ... ..."

"... In PAGE 74: ...74 Table 10: The number of misclassifications. Table11 illustrates the main benefit of the fuzzy classifier. The samples can simultaneously belong to several classes.... ..."

"... In PAGE 3: ... On the contrary, if THD is high and H19, H21, overvoltages are allowed - normal operation is expected. Investigation results are shown in Table4 . As before, most cases were correctly recognized, accordingly to the learning patterns (Table 4, cases 1.... In PAGE 3: ... Investigation results are shown in Table 4. As before, most cases were correctly recognized, accordingly to the learning patterns ( Table4 , cases 1.... ..."

"... In PAGE 9: ... The results of the fourth experiment with neuro-fuzzy model. When analysing Table8 the figures are really disgusting, especially for congested conditions. For all 9333 testing cases we have 95.... ..."