Abstract
Semantic Role Labeling (SRL) has proved to be a valuable tool for performing automatic analysis of natural language text. SRL requires additional information besides lexical and syntactical features to recognize the candidate arguments better. This paper investigates the use of Lexical Relation Pair (LRP) information for the task of Event Construction Extraction. Experimental results show that introducing LRP features can achieve some improvements. Further experiments indicate that using automatically extracted LRPs can obtain even better performance than manually annotated ones. This observation enables the large-scale usage of LRPs in the future.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have