### Table 2. Results from the weighted mean and Choquet integral aggregation operators for correlated criteria, particularized for c3 =0,c4 =0

2003

"... In PAGE 9: ... Of course, this direct elicitation approach is only feasible for decision problems with a small number of criteria, for which the number of mapping in the fuzzy measure is reasonably small. Table2 depicts an excerpt of the results from the weighted mean (w-mean) and Choquet integral (c-int) when the contribution of interactivity level (c3)and cost (c4) is void. Only integer values for inputs are shown, but in actual coordina- tion processes they may come from estimation processes not necessarily constrained to a small and finite number of levels.... ..."

### Table 3. Results from the weighted mean and Choquet integral aggregation operators for correlated criteria, particularized for c1 =3,c3 =0,c4 =0

2003

"... In PAGE 9: ... However, this phenomenon, caused by the positive correlation between both crite- ria, does not occur with the Choquet integral. Table3 points out a second example of the power of the Choquet integral to... ..."

### Table 2: Veto and favor indices for various Choquet integrals

2004

"... In PAGE 15: ... (24) shows that if both veto(Cv; j) and favor(Cv; j) increase then so does the importance `(v; j). Table2 gives the veto and favor indices for particular Choquet integrals.... ..."

Cited by 3

### Table 7. Classical description of intuitionistic binary clause entailment

1998

"... In PAGE 10: ... In these counter-models the world v is constrained so as to demonstrate the invalidity of a sequent: v forces all the premiss formulae of the sequent but not its conclusion. The conditions under which a binary formula of I is forced at v in Nebel apos;s counter- models can be speci ed classically as given in Table7 . To test if a sequent is valid we consider a set of constraints consisting of the forcing constraint for each premiss formula, the negation of the forcing constraint of the conclusion formula and also all instances of F(v; x) ! (F(w1; x) ^ F(w2; x)) (where x is a any constant occurring in the sequent), which arise from the ordering conditions on the worlds.... ..."

Cited by 7

### Tables 2-4 summarise the results of the experiments. In the first set of experiments we trained 20 neural networks with half amp;half sampling technique and combined them using the majority vote, the averaging, and the Choquet integral based technique with one common for the entire data space l -fuzzy measure. For this round of tests, the data space was not partitioned (column named K=1). The following notations are used in the tables: Mean stands for the percentage of the average classification error, Std is the standard deviation of the error, The Best stands for the single neural network with the best average performance, MV means majority vote, AV stands for the

### Table 5. Correlation Coefficients for Aqueous Solubility Blind Test Regressionsa

1998

"... In PAGE 9: ... Second, the calculated coefficients of the descrip- tors for the two group regressions are all within the error estimate of the coefficients for the regression made with all structures, suggesting that the coefficient values are reliable. Finally, when considering the correlation coefficients of the predictions for the blind tests in Table5 , it is apparent that the ability of the regressions made using two-thirds of the structures to predict the aqueous solubility for the excluded third is essentially equal. The average correlation coefficient for the blind cases (AB f C, AC f B, BC f A) was equal to the correlation coefficient (R2 ) 0.... ..."

Cited by 2

### Table 2: Correlation Coefficient between the frequency of accesses to a document and its size.

1999

"... In PAGE 11: ... Furthermore, across n, there does not seem to be strong correlation between document size and n, though the average size of popular documents is smaller than that of unpopular documents. We also calculated the correlation coefficient between the access frequency and document size for the traces, shown in Table2 . The numbers show that the correlation, if any, is weak and can be ignored.... ..."

Cited by 601