Point the Point: Uyghur Morphological Segmentation Using PointerNetwork with GRU

Yaofei Yang, Shupin Li, Yangsen Zhang, Hua Ping Zhang*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Citations (Scopus)

Abstract

Uyghur is an agglutinative language that has many morphemes. It is necessary for processing Uyghur to segment words into morphemes. This work is called morphological segmentation. Previous works treat morphological segmentation as a tagging task and classify each character as one of four classes, which are However, these labels are not independent from each other, which makes the models easily overfitted. We propose a new method for the segmentation task. Instead of using these labels, we use only segmentation points for modeling. The model used in our method is more robust and easier to train than previous methods. Applying our model to Uyghur morphological segmentation, it achieves high accuracy and higher recall and f1 score than previous models.

Original languageEnglish
Title of host publicationChinese Computational Linguistics - 18th China National Conference, CCL 2019, Proceedings
EditorsMaosong Sun, Yang Liu, Zhiyuan Liu, Xuanjing Huang, Heng Ji
PublisherSpringer Science and Business Media Deutschland GmbH
Pages371-381
Number of pages11
ISBN (Print)9783030323806
DOIs
Publication statusPublished - 2019
Event18th China National Conference on Computational Linguistics, CCL 2019 - Kunming, China
Duration: 18 Oct 201920 Oct 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11856 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference18th China National Conference on Computational Linguistics, CCL 2019
Country/TerritoryChina
CityKunming
Period18/10/1920/10/19

Keywords

  • Agglutinative language
  • Linguist
  • Morphological segmentation
  • NLP
  • PointerNetwork
  • Uyghur

Fingerprint

Dive into the research topics of 'Point the Point: Uyghur Morphological Segmentation Using PointerNetwork with GRU'. Together they form a unique fingerprint.

Cite this