### Table 3: Number of locations identified which are region members and ambiguous (using full IE)

2004

"... In PAGE 14: ... Table3 summarizes the locations found using the best method in Table 2 (full IE using additional grammar rules). Many of the locations found occur multiple times; therefore to obtain a more accurate view of the grounding we count multiple occurrences once (unique).... In PAGE 14: ... Many of the locations found occur multiple times; therefore to obtain a more accurate view of the grounding we count multiple occurrences once (unique). The second column in Table3 shows the number of unique locations extracted using the geo-parser. Many of these locations, however, cannot be grounded using the SPIRIT or OS resources.... In PAGE 14: ... The number of unique locations found is much smaller than the total number found (C+PC+FP) because many locations occur more than once (particularly in Wales and the Midlands). The third column in Table3 shows the number of unique locations grounded.... In PAGE 14: ... The fifth column shows the number of locations which are region members and have been grounded correctly (judged manually). The final column in Table3 shows the number of ambiguous locations and the proportion of these disambiguated correctly.... ..."

Cited by 13

### Table 3: Number of locations identified which are region members and ambiguous (using full IE)

"... In PAGE 9: ...ap24.com and Multimap (http://www.multimap.com) to visualize locations. Table3 summarizes the locations found using the best method in Table 2 (full IE using additional grammar rules). Many of the locations found occur multiple times; therefore to obtain a more accurate view of the grounding we count multiple occurrences once (unique).... In PAGE 9: ... Many of the locations found occur multiple times; therefore to obtain a more accurate view of the grounding we count multiple occurrences once (unique). The second column in Table3 shows the number of unique locations extracted using the geo-parser. Many of these locations, however, cannot be grounded using the SPIRIT or OS resources.... In PAGE 9: ... The number of unique locations found is much smaller than the total number found (C+PC+FP) because many locations occur more than once (particularly in Wales and the Midlands). The third column in Table3 shows the number of unique locations grounded.... In PAGE 10: ... The fifth column shows the number of locations which are region members and have been grounded correctly (judged manually). The final column in Table3 shows the number of ambiguous locations and the proportion of these disambiguated correctly.... ..."

### Table 2: The computed clusters and the associated categories

2007

"... In PAGE 11: ... In general, the clustering performance presents interesting results: the computed values in terms of recall and precision are rather accurate for the most of experiments. Table2 shows the main categories elicited by this experimentation: according to the topics of terms in the reference set, a pertinent category is associated. Furthermore, for each category, the more relevant sample of documents associated to the flrst three page links of returned list is shown.... ..."

### Table 1: Inner representation of expressions Term/Expression Inner Representation

"... In PAGE 2: ... The Domain Expert module con- verts string equations to their corresponding inner representations. Table1 represents corresponding in- ner representations of terms and expressions. Some examples of equations and their correspond- ing representations are provided by Table 2.... ..."

### Table 3. Lists of top collocates for the term gene sorted by MI and Z statistics

2000

Cited by 3

### Table 2 Piecewise Polynomial; contribution of control lines

"... In PAGE 6: ... Similarly, as the number of sections increases, the improvement resulting from the addition of another section diminishes. The next experiment was designed to investigate the contribution of linear features for a varying number of control points with the piecewise polynomial model; see Table2 . Adding straight line features to the solution generally improves the results, particularly when there are fewer control points.... ..."

### Table TERMS

1998

Cited by 1

### Table 1: Terms and their interpretation

1989

"... In PAGE 8: ...Extending the language In this section we discuss how to interpret terms with any nite number of variables (instead of exactly one as in Table1 ) and how datatypes relate to computations. We will consider only product and functional types, because sum types are completely straightforward5.... In PAGE 8: ... If T were IdC, then [[x1: 1 ` (let x2=e2 in e): ]] would be hid 1; g2i; g. In the general case, Table1 says that ; above is indeed composition in the Kleisli category, therefore hid 1; g2i; g becomes hid 1; g2i; T g; . But in hid 1; g2i; T g; there is a type mismatch, since the codomain of hid 1; g2i is 1 T 2, while the domain of T g is T ( 1 2).... In PAGE 18: ...SYNTAX SEMANTICS var x1; : : : ; xn ` xi = n i ; V let x ` e1 = g1 x; x ` e2 = g2 x ` (let x=e1 in e2) = hidV n; g1i; tV n;V ; T g2; V x; x ` e = g x ` ( x:e) = T V;V;V n(g); G; V V T app x ` e1 = g1 x ` e = g x ` e(e1) = hg; g1i; appv Table 9: call-by-value interpretation RULE SYNTAX SEMANTICS var x1; : : : ; xn ` xi = n i let x ` e1 = g1 x; x ` e2 = g2 x ` (let x=e1 in e2) = hid(TN)n; g1i; t(TN)n;N; T (id(TN)n N); T g2; N x; x ` e = g x ` ( x:e) = T TN;N;(TN)n(g); G; NTN T app x ` e1 = g1 x ` e = g x ` e(e1) = hg; g1i; appn Table1 0: call-by-name interpretation... ..."

Cited by 369