Author Description

This book proposes a linguistically motivated approach to terminological processing.Specifically, it investigates how syntactic functions can be utilized for enhanced term extraction. The rationale of this work stems from the understanding that terms tend to perform particular syntactic functions more prominently and their syntactic behaviour can be captured and represented as termhood by computing term ratios in different syntactic paths. In essence, this method is a weighting scheme that measures the probabilistic relations between term occurring patterns and syntactic paths. The studies presented in this book are aimed at answering several research questions: whether syntactic information is useful for recognizing candidate terms in medical text, how this syntactically oriented metric can be used to improve the ranking of multi-word terms, and whether this metric can be used effectively for novel term recognition as an effective feature in a machine learning paradigm. This research has led to an implemented system which demonstrates superior performance when compared with some major publically accessible systems for term extraction.