### Table 3. Rule Set for Fuzzy Logic Controller

2003

"... In PAGE 8: ...able 2: Example of Driver I/O .........................................................................................30 Table3 : Rule Set for Fuzzy Logic Controller.... In PAGE 50: ... Once the input is fuzzified and all the output fuzzy sets are defined as appropriate linguistic terms, the fuzzy inference engine looks for a match between the input and the outputs. The agent has 17 fuzzy rules in total: seven rules for the front sensors, two for the rear sensors, and four for each of the left and right sensors as shown in Table3 . Left columns in the table correspond the antecedents of the fuzzy rules.... ..."

### Table 1 Fuzzy rules

"... In PAGE 9: ... Based on the assumption of input and output parameters, the fuzzy rule FR in our fuzzy logic controller is IF MinSupport is A and Lean is B THEN RealSupport is C where A, B and C are fuzzy sets. The following Table1 is an example for illustrating the construction of fuzzy rules. In Table 1, the first column is the fuzzy sets in F Lean; the first row is the fuzzy sets in F MinSupport; and others are the outputs generated for Real- Support.... In PAGE 9: ... The following Table 1 is an example for illustrating the construction of fuzzy rules. In Table1 , the first column is the fuzzy sets in F Lean; the first row is the fuzzy sets in F MinSupport; and others are the outputs generated for Real- Support. Each output is a fuzzy rule.... ..."

### Table 4: Fuzzy query evaluation in the testing set

"... In PAGE 19: ... The KFC-L showed the best performance for the image data set, but for the ringnorm data set, the KFC-L showed a large error compared to other methods. Table4 shows the best recognition rates of the ringnorm validation data sets. The recognition rates of the KFC-L are very low com- pared to those of the KFC.... In PAGE 20: ...Table4 . Cross validation results (%) for the ringnorm data set.... In PAGE 29: ... The Precision and Recall metrics computed by the J4.8 algorithm over the same testing set are also shown in Table4 . It can be seen that the fuzzy query results over the testing set correspond roughly to the ones computed from the training set.... In PAGE 48: ...ig. 7. Comparison between heuristic rule extraction and genetics- bases multiobjective rule selection with respect to the average error rates on training patterns (Wine). Wisconsin Breast Cancer Data: Experimental results by heuristic rule extraction are summarized in Table4 where the average error rate on test patterns over three iterations of the 10CV procedure is shown for each combination of a heuristic rule extraction criterion and the number of extracted rules for each class. The best error rate in each row is indicated by bold face.... In PAGE 48: ...5 algorithm with respect to the generalization ability. Table4 . Average error rates on test patterns of extracted rules by heuristic rule extraction (Breast W).... In PAGE 86: ...007 isdn .009 Table4 . Continuation of onramp/offramp traf- fic distribution for Figure 5.... ..."

### Table 2. Fuzzy rules of torque hysteresis controller

"... In PAGE 3: ... The shapes of mem- bership functions are refined trough simulation and test- ing. The rules sets are shown in Table2 . Figure 5 shows the membership functions of input and output variables.... ..."

### Table 1 presents a classi cation of the bibliography on the combination of GAs and FL collected in [1]. It contains the keywords and the number of papers on each of them. With these keywords the application of FL based tools to GA (with the name of fuzzy genetic algorithms) and the di erent areas of the fuzzy logic and fuzzy sets theory, [6], where GAs have been applied are covered. 1 Fuzzy genetic algorithms 16

1996

"... In PAGE 1: ... Table1 . Classi cation keywords As it can be noted at the sight of the big quantity of papers in the area, the most widely studied problem for which GAs have proven very useful is the design, learning and tuning of fuzzy logic controllers (FLCs).... In PAGE 3: ... In particular, the application to the design, learning and tuning of fuzzy control rules has given quite promising re- sults. This application approach has been widely studied as we can see in the Table1 where approximately half of references is in this topic. Due to the lack of space and the importance of this application, we will focus in its study in this section.... ..."

Cited by 6

### Table 1 presents a classi cation of the bibliography on the combination of GAs and FL collected in [1]. It contains the keywords and the number of papers on each of them. With these keywords the application of FL based tools to GA (with the name of fuzzy genetic algorithms) and the di erent areas of the fuzzy logic and fuzzy sets theory, [6], where GAs have been applied are covered. 1 Fuzzy genetic algorithms 16

1996

"... In PAGE 1: ... Table1 . Classi cation keywords As it can be noted at the sight of the big quantity of papers in the area, the most widely studied problem for which GAs have proven very useful is the design, learning and tuning of fuzzy logic controllers (FLCs).... In PAGE 3: ... In particular, the application to the design, learning and tuning of fuzzy control rules has given quite promising re- sults. This application approach has been widely studied as we can see in the Table1 where approximately half of references is in this topic. Due to the lack of space and the importance of this application, we will focus in its study in this section.... ..."

Cited by 6

### Table 1. Fuzzy control rules

"... In PAGE 3: ... 4. Simulation Because the linguistic terms, such as positive big (PB), positive medium (PM), positive small (PS), zero (ZE), negative small (NS), negative medium (NM), negative big (NB), are used, there are 49 rules ( see Table1 ) in the system. Table 1.... ..."

### Table 4. Fuzzy control rule

"... In PAGE 13: ...In Table4 , NL means #5CNegatively Large quot;, NM means #5CNegatively Medium quot;, NS means #5CNega- tively Small quot;, ZE means #5CZero Equivalence quot;, PS means #5CPositively Small quot;, PM means #5CPositively Medium quot;, PL means #5CPositively Large quot;, S means #5CSmall quot;, M means #5CMedium quot;, and L means #5CLarge quot; #28Kosko, 1992, and Zurada, 1992#29. Note that this rule could be generic for all neural networks using back-propagation learning algorithms.... ..."