Increasing NLP Parsing Efficiency with Chunking

Proceedings

Vilares, David¹  Anderson, Mark Dáibhidh² 

[1] Author to whom correspondence should be addressed.;FASTPARSE Lab, Departamento de Computación, University of A Coruña, Campus de Elviña, 15071 A Coruña, Spain

关键词: Parsing;    Syntax;    natural language processing;    NLP;    dependency parsing;    Chunking;   

DOI  :  10.3390/proceedings2181160

学科分类：社会科学、人文和艺术（综合）

来源: mdpi

【 摘 要 】

We introduce a “Chunk-and-Pass” parsing technique influenced by a psycholinguistic model, where linguistic information is processed not word-by-word but rather in larger chunks of words. We present preliminary results that show that it is feasible to compress linguistic data into chunks without significantly diminishing parsing performance and potentially increasing the speed.

【 授权许可】

CC BY