Development of a Conceptual Structure for a Domain-Specific Corpus
This work addresses the problem of domain-specific knowledge extraction for researchers in natural language processing, but it is incremental as it builds on existing methods for corpus development.
The paper tackled the challenge of identifying and extracting semantic relations from a domain-specific corpus for a Text to Knowledge Mapping prototype, resulting in a manually developed conceptual structure that produced a framework of 55 semantic relations, with 42 having inverse relations.
The corpus reported in this paper was developed for the evaluation of a domain-specific Text to Knowledge Mapping (TKM) prototype. The TKM prototype operates on the basis of both a combinatory categorical grammar (CCG) linguistic model and a knowledge model that consists of three layers: ontology, qualitative and quantitative layers. In the course of this evaluation it was necessary to populate these initial models with lexical items and semantic relations. Both elements, the lexicon and semantic relations, are meant to reflect the domain of the prototype; hence both had to be extracted from the corpus. While dealing with the lexicon was straight forward, the identification and extraction of appropriate semantic relations was much more involved. It was necessary, therefore, to manually develop a conceptual structure for the domain which was then used to formulate a domain-specific framework of semantic relations. The conceptual structure was developed using the Cmap tool of IHMC. The framework of semantic relations- that has resulted from this study consisted of 55 relations, out of which 42 have inverse relations.