### Table 4. Mapping of SQL Statements to Equiv- alent Persistence Methods

1996

Cited by 4

### Table 4. Mapping of SQL Statements to Equiv- alent Persistence Methods

1996

Cited by 4

### Table 3. Comparison of Slab and CA Dose-Equiv alent Estimates

"... In PAGE 25: ...Table3 . Comparison of Slab and CAM Dose-Equivalent Estimates #5BData from #28ref.... ..."

### Table 4: Strengths of the interior and simplex CP approaches after an equiv- alent no of iterations

### Table 1. The semantic value of an expression E in an interpretation I is the set EI; two expressions are equiv- alent if they have equal semantic values in every inter- pretation.

1997

"... In PAGE 1: ...xist. quant. (E) 9R:C fd1 j 9d2 (d1; d2) 2 RI ^ d2 2 CIg number ( nR) fd1 j jf(d1; d2) 2 RIgj ng restriction (N) ( nR) fd1 j jf(d1; d2) 2 RIgj ng role name R RI I I role conj. (R) Q u R QI \ RI Table1 : Constructors in First-Order Description Logics We refer the reader to [Donini et al., 1996] for a recent overview of the area.... In PAGE 2: ... Let us consider each of these in turn, starting with item I. As is well-known, each extension of FL? that is de ned using the constructors from Table1 can be viewed as a fragment of rst-order logic over a suitable vocabulary; thus, in this paper we use rst-order logic as our background logic. Item II is the heart of the method.... In PAGE 2: ... Item II is the heart of the method. With each fragment corresponding to a description logic L that is de ned us- ing the constructors in Table1 we will associate an L- relation between interpretations that characterizes L in the following way. Roughly speaking, a rst-order for- mula is (equivalent to) an L-concept if, and only if, it re- mains true under passing from an interpretation to an L- related interpretation.... In PAGE 2: ... 4 Main Results Our results come in two kinds. First, we establish the model-theoretic characterizations announced above for every extension of FL? and AL that can be de ned us- ing the constructors from Table1 . Second, we use these characterizations to separate description logics.... In PAGE 3: ... A R R R t t PPPPq 1 t t t d1 d2 I J Figure 1: Separating FL? and FLE? The above separation result is an instance of a more general result which completely classi es the expressive power of all description logics extending FL? or AL. Theorem 3 (Classi cation) Using the characteriza- tions of Theorem 2 all extensions of FL? and AL ob- tained by adding the constructors in Table1 can be com-... ..."

Cited by 7

### Table 1. The semantic value of an expression E in an interpretation I is the set EI; two expressions are equiv- alent if they have equal semantic values in every inter- pretation.

1997

"... In PAGE 1: ...xist. quant. (E) 9R:C fd1 j 9d2 (d1; d2) 2 RI ^ d2 2 CIg number ( nR) fd1 j jf(d1; d2) 2 RIgj ng restriction (N) ( nR) fd1 j jf(d1; d2) 2 RIgj ng role name R RI I I role conj. (R) Q u R QI \ RI Table1 : Constructors in First-Order Description Logics We refer the reader to [Donini et al., 1996] for a recent overview of the area.... In PAGE 2: ... Let us consider each of these in turn, starting with item I. As is well-known, each extension of FL? that is de ned using the constructors from Table1 can be viewed as a fragment of rst-order logic over a suitable vocabulary; thus, in this paper we use rst-order logic as our background logic. Item II is the heart of the method.... In PAGE 2: ... Item II is the heart of the method. With each fragment corresponding to a description logic L that is de ned us- ing the constructors in Table1 we will associate an L- relation between interpretations that characterizes L in the following way. Roughly speaking, a rst-order for- mula is (equivalent to) an L-concept if, and only if, it re- mains true under passing from an interpretation to an L- related interpretation.... In PAGE 2: ... 4 Main Results Our results come in two kinds. First, we establish the model-theoretic characterizations announced above for every extension of FL? and AL that can be de ned us- ing the constructors from Table1 . Second, we use these characterizations to separate description logics.... In PAGE 3: ... A R R R t t PPPPq 1 t t t d1 d2 I J Figure 1: Separating FL? and FLE? The above separation result is an instance of a more general result which completely classi es the expressive power of all description logics extending FL? or AL. Theorem 3 (Classi cation) Using the characteriza- tions of Theorem 2 all extensions of FL? and AL ob- tained by adding the constructors in Table1 can be com-... ..."

Cited by 7

### Table 3 gives an overview of several relative quality indicators from the lit- erature. For each preference relation on approximation sets it depicts the equiv- alent conditions on the indicator values, if such conditions are possible at all.

1973

"... In PAGE 109: ... Also, the SPEA2 working with the floating point representation does not cover much (less than 10%) of the so- lutions produced by the integer version, which in turn is able to dominate nearly half of the solutions of its competitor. However, as far as the preference relations on approximation sets are concerned, the relations listed in Table3 lead to the... ..."

Cited by 1

### Table 1: Confusion matrix for flow content characterization using payload of size 16384 bytes (or roughly equiv- alent to using payloads from 16 packets).

### Table 2: Mean absolute deviation between true and recovered signal, for various filters de- scribed in the text. The complexity represents the number of Kalman filter updates (or equiv- alent) required by each filter at each time point, and is linearly related to speed.

1999

"... In PAGE 14: ... Parameter pJ w v h g Value 1=20 p5 100 p0:05 100 Table 1: Parameter values used to generate Figure 6.3 The results are shown in Table2 which gives the mean absolute deviation between the mean of the posterior distribution at time t and the true signal at time t: For this example, the simple SIR filter always diverges before the end of this series, even with sample sizes as large as 20; 000: This is because there always comes a time when the signal... ..."

Cited by 20

### Table 2 shows some basic document collection statistics. Note that although the collection sizes are roughly equiv- alent in megabytes, there is a range of document lengths across collections, from very short documents (DOE) to very long (FR). Also the range of document lengths within a collection varies. For example, the documents from AP are similar in length (the median and the average length are very close), but the WSJ, ZIFF and especially FR documents have much wider range of lengths within their collections.

"... In PAGE 3: ...Table2 : Document Statistics Subset of collection WSJ (disks 1 and 2) AP ZIFF FR (disks 1 and 2) DOE SJMN (disk 3) PAT(disk 3) Size of collection (megabytes) (disk 1) 270 259 245 262 186 (disk 2) 247 241 178 211 (disk 3) 290 242 349 245 Number of records (disk 1) 98,732 84,678 75,180 25,960 226,087 (disk 2) 74,520 79,919 56,920 19,860 (disk 3) 90,257 78,321 161,021 6,711 Median number of terms per record (disk 1) 182 353 181 313 82 (disk 2) 218 346 167 315 (disk 3) 279 358 119 2896 Av erage number of terms per record (disk 1) 329 375 412 1017 89 (disk 2) 377 370 394 1073 (disk 3) 337 379 263 3543 distinct parts to this collection -- the documents, the ques- tions or topics, and the relevance judgments or quot;right answers. quot; These test collection components are discussed very briefly in the rest of this section.... ..."