Close menu

SURE

Sunderland Repository records the research produced by the University of Sunderland including practice-based research and theses.

Deep Learning for Natural Language Parsing

This is the latest version of this item.

Jaf, Sardar and Calder, Calum (2019) Deep Learning for Natural Language Parsing. IEEE Access, 7 (1). pp. 131363-131373. ISSN 2169-3536

Item Type: Article

Abstract

Natural language processing problems (such as speech recognition, text-based data mining, and text or speech generation) are becoming increasingly important. Before effectively approaching many of these problems, it is necessary to process the syntactic structures of the sentences. Syntactic parsing is the task of constructing a syntactic parse tree over a sentence which describes the structure of the sentence. Parse trees are used as part of many language processing applications. In this paper, we present a multi-lingual dependency parser. Using advanced deep learning techniques, our parser architecture tackles common issues with parsing such as long-distance head attachment, while using ‘architecture engineering’ to adapt to each target language in order to reduce the feature engineering often required for parsing tasks. We implement a parser based on this architecture to utilize transfer learning techniques to address important issues related with limited-resourced language. We exceed the accuracy of state-of-the-art parser on languages with limited training resources by a considerable margin. We present promising
results for solving core problems in natural language parsing, while also performing at state-of-the-art
accuracy on general parsing tasks.

[img]
Preview
PDF
FINAL VERSION.pdf - Accepted Version

Download (631kB) | Preview

More Information

Related URLs:
Depositing User: Sardar Jaf

Identifiers

Item ID: 11185
Identification Number: https://doi.org/10.1109/ACCESS.2019.2939687
ISSN: 2169-3536
URI: http://sure.sunderland.ac.uk/id/eprint/11185
Official URL: https://ieeeaccess.ieee.org/

Users with ORCIDS

ORCID for Sardar Jaf: ORCID iD orcid.org/0000-0002-5620-0277

Catalogue record

Date Deposited: 10 Oct 2019 09:27
Last Modified: 30 Sep 2020 11:19

Contributors

Author: Sardar Jaf ORCID iD
Author: Calum Calder

University Divisions

Faculty of Technology
Faculty of Technology > School of Computer Science

Subjects

Computing > Artificial Intelligence
Languages > Languages
Computing > Programming
Computing > Software Engineering
Computing

Actions (login required)

View Item View Item

Available Versions of this Item