### Table 6.1: Part-whole Relations with their characteristics

### Table 1. Classification of the Part-Whole Relations from [28] w.r.t. our Taxonomy.

2000

Cited by 15

### Table 1: The list of all the constrains accompanied by examples (! means is not , 1 means Is a part-whole relation, and 0 means Is not a part-whole relation )

2003

"... In PAGE 5: ... 3.5 The Constraints Table1 summarizes the constraints learned by the pro- gram. The meaning of a constraint with the part Class-X, the whole Class-Y and the value 1 is if Part is a Class-X and Whole is a Class-Y then it is a part-whole relation and for the value 0 is if Part is a Class-X and Whole is a Class-Y then it is not a part-whole relation .... ..."

Cited by 26

### Table 1: The list of all the constrains accompanied by examples (! means is not , 1 means Is a part-whole relation, and 0 means Is not a part-whole relation )

2003

"... In PAGE 5: ... 3.5 The Constraints Table1 summarizes the constraints learned by the pro- gram. The meaning of a constraint with the part Class-X, the whole Class-Y and the value 1 is if Part is a Class-X and Whole is a Class-Y then it is a part-whole relation and for the value 0 is if Part is a Class-X and Whole is a Class-Y then it is not a part-whole relation .... ..."

Cited by 26

### Table 3. Average pattern performance per pattern class. T is the number of times patterns in the class occur in the training set. D is the number of discovered part-whole phrases.

2006

"... In PAGE 5: ...ig.2. (left) Frequency distribution in the training set of the learnt patterns. Referred to as T in Table3 .... In PAGE 6: ... Table 2 shows that together, the implicit patterns account for a third of the total number of part-whole pairs. When applying patterns to learn part-whole relations it is useful to make this dis- tinction into three types, because it turns out that these three types have rather different Precision and Recall properties, listed in Table3 . The patterns in the part of class yield the most results with high Precision.... ..."

Cited by 4

### Table 1: Part-whole relationships.

1994

Cited by 1

### Table 2: Failure analysis of spurious part-whole relations found by PartEx, for configuration BASE (over half the spurious relations across ten folds) and configuration FILT (all spurious relations in ten folds). In each case, a small number of relations are in two categories.

"... In PAGE 5: ... Improvements could also be made with more sophis- ticated context, and by examining compounds. A failure analysis of spurious relations for config- uration FILT is shown in Table2 . The biggest im- pact on precision was made by relations that could be considered correct, as discussed in Section 4.... ..."

### Table 6. The 91 patterns used for the learning of wholes, ordered by the number of correct pairs it yielded. Prec. is Precision approximated on a sample of 25 occurrences (or less if freq. lt; 25). D is the number of discovered part-whole phrases.

2006

"... In PAGE 7: ...7,357 labels in total. Thus limiting the set of discoverable wholes to 69,746 concepts. For each remaining whole in the list we construct a part-whole relation. An assessment of the part-whole results is shown in Table6 . We approximated Precision for the 91 patterns we used to find wholes based on a random sample of 12 http://www.... ..."

Cited by 4

### Table 4b: Part-Whole Reasoning - (ii) concept specification

"... In PAGE 8: ... Trans Instance level: FORALL C,D,X,Y,Z,S,R X[R- gt; gt;Z] lt;- PartonomicRolePropagationOn(C,R,D) and X:C and Y:D and X[R- gt; gt;Y] and subrelationOf(S,partOf) and Y[S- gt; gt;Z]. Concept level: analogous to the instance level Table4 a: Part-Whole Reasoning - (i) role propagation The second aspect of partonomic reasoning that is reflected upon by Rector et al. is concept specification through partonomic reasoing.... In PAGE 8: ... is concept specification through partonomic reasoing. To cut it short, we only mention here that this type of axiom specification also relies on the PartonomicRolePropagation specified in Table4 a and show... In PAGE 8: ...Table4... ..."