Knowledge Representation and Metaphor
Knowledge Representation and Metaphor
| Hi-index | 0.00 |
The objective of this study is to show how the contextual knowledge about the topic allows choosing one out of the many interpretations possible for attributive metaphors of the form [A (the topic) is B (the vehicle)]. We propose an experiment that shows (i) that the properties of the vehicle that are actually attributed to the topic are the ones that are specified by the context in which the topic appears; and (ii) that conventional interpretations of a metaphor do not predominate unless this context is neutral, i.e. when the attributes of the topic furnished by the context do not belong to the vehicle's category. We interpret these results on the importance of the context in understanding metaphor both in terms of property matching [1, 4, 7] and in terms of categorization [2, 8]. In our approach, understanding a metaphor consists of including the topic in a category specified by the vehicle such that the topic inherits certain of the traits of this category, within a situational object category network.