| ||Format||Pages||Price|| |
|PDF (340K)||19||$25||  ADD TO CART|
|Complete Source PDF (3.9M)||247||$93||  ADD TO CART|
In this paper, we describe an approach to computer-assisted definition construction using CODE, a knowledge management system developed at the Artificial Intelligence Lab, Department of Computer Science, University of Ottawa, Canada. Our approach, which we call definition-oriented concept analysis, tries to rectify some of the most common problems in intensional terminological definitions, namely inconsistency of genus and differentia from one coordinate concept to the next. We examine how various features of CODE can help terminologists carry out definition-oriented concept analysis. These features include textual and graphical browsers, along with the Characteristic Comparison Matrix (CCM), a special browser that lets users view and compare the characteristics of coordinate concepts in tabular form. The explanation of our approach includes practical examples taken from a terminological knowledge base (TKB) on lasers, and demonstrates how terminologists can use CODE to methodically carry out definition construction and create consistent definitions.
definition, intensional definition, computer-assisted definition construction, definition-oriented concept analysis, artificial intelligence, computational terminology, knowledge engineering, knowledge management
Senior Information Developer, Bell-Northern Research Ltd., Ottawa, Ontario
Associate Professor, School of Translation and Interpretation, 52 University/Priv., University of Ottawa, Ottawa, Ontario