A Smote-Based News-Specific geographical name extraction and classification algorithm

Chen Zhang, Chongyang Shi

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

Abstract

A geographical name is a conventional language sign, which is not only used for a geographic entity, but also widely applied in the naming of other entities, thus significantly affecting the accuracy of news-specific geographical name extractions. In this paper, the characteristics of the geographical names in the news are analyzed and the geographical names are divided into geographic entity, organizational entity, in addition to other entities. For the unbalance of these three entities in the news, a SMOTE-based random walk algorithm is proposed here to realize the classification and identification of geographical names.

Original languageEnglish
Title of host publicationProceedings of the IASTED International Conference Modelling, Identification and Control, MIC 2017
PublisherActa Press
Pages101-105
Number of pages5
ISBN (Electronic)9780889869882
DOIs
Publication statusPublished - 2017
Event2017 IASTED International Conference on Modelling, Identification and Control, MIC 2017 - Innsbruck, Austria
Duration: 20 Feb 201721 Feb 2017

Publication series

NameProceedings of the IASTED International Conference on Modelling, Identification and Control
Volume848
ISSN (Print)1025-8973

Conference

Conference2017 IASTED International Conference on Modelling, Identification and Control, MIC 2017
Country/TerritoryAustria
CityInnsbruck
Period20/02/1721/02/17

Keywords

  • Entity extraction
  • Geographical Name Classification
  • Geographical name Extraction
  • SMOTE

Fingerprint

Dive into the research topics of 'A Smote-Based News-Specific geographical name extraction and classification algorithm'. Together they form a unique fingerprint.

Cite this