Context and Terminology
Pamela Faber (
pfaber <at> ugr (point) es)
LexiCon Research Group, Universidad de Granada
Le 10 juillet 2013 à 11 h 30
Salle C-9019, Pavillon Lionel-Groulx
A well-structured terminological knowledge base (TKB) provides an access point to domain knowledge and greatly facilitates knowledge acquisition. Consequently, the design decisions for a TKB should not be made randomly, but rather should be firmly grounded in a coherent set of premises derived from linguistic theory and knowledge organization. These premises are the framework for corpus pattern analysis, the formulation of definitions, the specification of conceptual representations, and meaning contextualization. The selection of optimal design parameters is crucial because this decision affects the type and quantity of information in each term entry microstructure) and the way that entries are related to each other (macrostructure), even when this relational structure is not explicitly encoded in the TKB. Indeed, it is the structure and configuration of conceptual information even more than the sheer quantity of data that convert a database, of which there are many, into a knowledge base, of which there are relatively few. If the terminological knowledge base is multilingual, this adds still another level of complexity. The specification of a common conceptual structure applicable to and valid for different language-cultures requires a representational framework that allows for correspondences at different levels as well as for the inclusion of the conceptual-semantic and pragmatic features upon which correspondence is based. Also important is the syntactic information that languages use to encode the various types of conceptual relation that link concepts in different languages. In this regard, context is an important construct in the description of the concepts and terms of any specialized domain because word and term meaning are extremely context-sensitive. Although terminological knowledge bases may provide conceptual representations that are based on some sort of conceptual model and modeling mechanisms, they rarely support context, i.e. providing controlled partial information on conceptual entities by viewing them from different viewpoints or situations. This can be a problem because the meaning as well as the location of a concept within a knowledge structure often varies, depending on its context. Matters are further complicated by the fact that context itself is a complex entity, which can come in varying shapes and sizes. The lecture proposes an expanded classification of context in Terminology, which goes beyond the categories defined in previous research. This classification is based on the local or global scope of the context as well as the different types of semantic, syntactic, and/or pragmatic information contained in it.