### Table 1: The Strengths of Preferences and Matches

1994

"... In PAGE 3: ... For instance, in considering the attribute workload, the strength of the preference will be low-moderate, and the closeness of the match will be strong and exact for CS601 and CS621, respectively. Table1 shows a summary of the strength of the preferences and the closeness of the matches for the relevant attributes for both instantiations. Numerical values are then assigned and used to calculate a final weight for each candidate.... In PAGE 6: ...Table1 , which suggests that CS601 is a substantially better alternative to CS621. Thus, Improve-Parameter is selected as a special- ization of Modify-Proposal.... In PAGE 6: ... (8) CS601 meets at 2pm and involves formal lan- guages and grammar. Notice that utterance (8) provides supporting evidence for the claim in (7), and is obtained by comparing the sets of information used by the ranking advisor ( Table1 ) and selecting the features that contribute most to making CS601 preferable to CS621. The Belief Level We showed how our arbitrator detects and resolves con- flicts at the domain level.... ..."

Cited by 56

### Table 2: Determinants of the preferences for environmental protection in Spain

"... In PAGE 19: ... This section reports two groups of estimation results. Table2 presents baseline estimation checking the robustness of the results working with or without weighting variables. Furthermore, to reduce possible causality problems, 2SLS estimations are presented.... In PAGE 19: ... Female report a higher preference to contribute than men. The coefficient is statistically not significant in the non-weighted estimations, but significant in the first weighted estimations in Table2 and 3. Estimation 2 indicates that being female rather than male increases the probability of a person to strongly agree to increase taxes to prevent environmental damage by 1.... ..."

### Table 14: Stability for additional weight to downstream apron

"... In PAGE 8: ...able 13: Stability for additional weight on powerhouse, no earthquake ...............................48 Table14 : Stability for additional weight to downstream apron .... In PAGE 60: ... Appendix 19 shows the placement of additional walls 9 m wide walls of 25 m respectively 21 m height (jump in downstream apron causes difference in height wall) in section 6 and section 10 analyses the stability. Table14 shows the results of the stability analysis and provides a comparison with the situation in case the drainage system under the power house has failed entirely and the case in which the original design drainage is functional. This alternative does not provide sufficient stability for the power house when no earthquake is taking place.... ..."

### Table 13: Stability for additional weight on powerhouse, no earthquake

"... In PAGE 8: ...able 12: Allowed and actual distance of resulting force from middle effective width............43 Table13 : Stability for additional weight on powerhouse, no earthquake .... In PAGE 58: ... Walls 6 m wide and 15 m and 10 m high on the horizontal respectively sloping part of the spillway combined with a roof of 5 m on top of the walls on the horizontal part of the spillway were implemented in the calculation. Table13 shows the results of the stability analysis for additional walls and provides a comparison with the situation in case the drainage system under the power house has failed entirely and the case in which the original design drainage is functional. This alternative does not provide sufficient stability for the power house.... ..."

### Table 1: Existing computational complexity results for the stability semantics

2002

"... In PAGE 8: ... The complexity of the problems we are interested in has been extensively stud- ied for existing logics for default reasoning under the standard, stability seman- tics [3, 15, 23, 19, 29, 2, 10]. Table1 gives a partial summary of these results for the different logics. We note here that the semantics of circumscription has been originally proposed with respect to sceptical reasoning only.... In PAGE 19: ... Therefore, A9 is true. Therefore, as in the LP case, in the DL case credulous reasoning under the admissibility and preferability semantics has the same complexity as under the stability semantics (see Table1 ), whereas sceptical reasoning is either one level... ..."

Cited by 21

### Table 1 Comparison resutls among methods

"... In PAGE 4: ... [4] We compare the retrieval times for both sequential and parallel searches using one feature, with a multi-feature search using both a regular weighted algorithm and a weighted cascade. The quantitative results are shown in Table1 . Retrieval time and speed up are shown in Figure 5.... ..."

### Table 5: Preference weight assignment for pronominal anaphora

### Table 5 : bi-matching

"... In PAGE 11: ...Table 5 : bi-matching Table5 compares all four algorithms on two problems where the cost and weight functions are randomly generated with values ranging from 1 to 100. There is a balance between optimization on the cost and feasibility on energy : they both can cut the search tree.... ..."

### Table 6 Results for the Phase Stability Problem

"... In PAGE 22: ... These problems are taken from McDonald and Floudas (1995) and have been solved by them using the GLOPEQ package (McDonald and Floudas, 1994). The results are shown in Table6 . It can be seen that for most of the problems, the GOP algorithm performs very well when compared to the specialized code in GLOPEQ, which is a package specifically designed for phase equilibrium problems.... ..."

### Table 6 The stability metrics of 10000 non-match similarity scores.

in Using Chebyshev’s Inequality to Determine Sample Size in Biometric Evaluation of Fingerprint Data

2005

"... In PAGE 23: ... As a result, the case of 10000 non-match similarity scores is chosen to show the stability. The stability metrics from 100 to 500 Monte Carlo iterations for sample size of 10000 are presented in Table6 , for Matcher 1 and 2 as well as for two different criteria of AUROC and TVAFV, respectively. As expected, the stability metric of Matcher 1 is smaller than the one of Matcher 2 for a fixed criterion, and the stability metric of AUROC is smaller than the one of TVAFV for a specified matcher.... In PAGE 24: ...20 In Table6 , it shows that the outcome of Monte Carlo calculation for 10000 non-match similarity scores is very stable from 100 iterations up to 500 iterations with respect to specified matcher and criterion. The worst deviations of the one-trial test result from the baseline result with greater-than-95% probability vary by no-larger-than 0.... In PAGE 24: ...reater-than-95% probability vary by no-larger-than 0.000015, 0.000061, 0.000224, and 0.003257 (the maximum minus the minimum in each row of Table6 ) from 100 to 500 iterations for Matcher 1 and 2 and for two different criteria, respectively, even when the sample size is down to only 10000 non-match similarity scores. As a consequence, the results presented above out of 500 Monte Carlo iterations are reliable.... ..."