Retrieving Collocation Frameworks for Entity Attribute Knowledge Acquisition

DOI

10.2991/fmsmt-17.2017.301How to use a DOI?

Keywords

Retrieving, Collocation, Entity Attribute.

Abstract

The key problem in the acquisition of the entity attribute knowledge for natural language understanding lies in the connections between the entity attributes. These connections could be represented by entity attribute collocations. It is impossible to get these entity attribute collocations manually. This paper proposed a method of retrieving collocation frameworks for entity attribute knowledge acquisition, which could acquire the entity attribute collocations from real corpus automatically. Because the collection framework template is actually the simplest syntactic sub-tree which retained the core verbs and the brother branch of the entity word and the attribute around the core verb. The proposed method obtained the entity attribute collocations based on the pruning of the syntactic tree. The experimental result showed that the proposed method performance well on the real corpus.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Hong-lin Wu
AU  - Ruo-yi Zhou
AU  - Ke Wang
PY  - 2017/04
DA  - 2017/04
TI  - Retrieving Collocation Frameworks for Entity Attribute Knowledge Acquisition
BT  - Proceedings of the 2017 5th International Conference on Frontiers of Manufacturing Science and Measuring Technology (FMSMT 2017)
PB  - Atlantis Press
SP  - 1550
EP  - 1553
SN  - 2352-5401
UR  - https://doi.org/10.2991/fmsmt-17.2017.301
DO  - 10.2991/fmsmt-17.2017.301
ID  - Wu2017/04
ER  -