Research
People
Software
Demos
Publications
Resources
Schedule
Login
Themes
|
Projects
|
Funding
|
Past Projects
Relation Extraction with Auxilliary Tasks
[
Overview
|
Participants
]
Overview:
Relation extraction (RE) is an important task in natural language processing. However, traditionally RE approaches use machine learning algorithms to predict the relationships between two different entities directly. In this project, we investigate the possibility of creating auxiliary tasks to improve the results. We hope that the auxiliary tasks can not only increase the precision of the learning models but also provide constraints to improve the final RE performance.
Participants:
Sander Canisius
,
Ming-Wei Chang
,
Dan Roth
Copyright © 2009,
University of Illinois at Urbana-Champaign
, All Rights Reserved.
