Knowledge Extraction on MultidimensionalConcepts: Corpus Pattern Analysis (CPA) andConcordances1

Terminology and Artificial Intelligence 2009.

计算机科学;

Pilar León Araúz ; Arianne Reimerink ; Pamela Faber2

Others  :  http://ceur-ws.org/Vol-578/paper24.pdf
PID  :  42709

学科分类：计算机科学（综合）

来源: CEUR

【 摘 要 】

Multidimensionality of concepts in multidisciplinary domains is a problem terminographers have to deal with. We apply Corpus Pattern Analysis(CPA; Pustejovsky, Hanks, & Rumshisky, 2004) to extract conceptualdimensions according to context. The dynamic nature of these concepts isexemplified with the case study of SAND. On the other hand, knowledgepatterns (KPs) often convey different conceptual relations and are thereforepolysemic structures. The development of pattern-based constraints can help todisambiguate them and at the same time avoid conceptual noise, which wouldbe a first step towards the systematization of automatic knowledge extraction.Two KPs are analyzed in detail: rang* from, which conveys the conceptualrelation is_a, and the polysemic KP formed by.