Climate event detection algorithm based on climate category word embedding

Hengyi Wang, Zhendong Niu

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

1 Citation (Scopus)

Abstract

Detecting climate events efficiently and accurately is important in traffic warning, disaster warning, and disease prevention. Given that ordinary event detection algorithms are ignored in climate domain's knowledge, the results of climate domain's event detection are not satisfactory. This paper proposes a climate event detection algorithm based on climate category word embedding(CEDCWE). This method combines the climate category word embedding and typical factors of climate events as a document representation model to express detailed information of climate documents and detect climate events efficiently and accurately. Compared with other methods, the CEDCWE algorithm can generate better climate document representations and climate event detection results. In the experiments, we acquire the datasets by a web crawler and evaluate our CEDCWE on real-world climate event detection tasks. Experimental results show that our CEDCWE is effective in climate document representation and outperforms typical methods.

Original languageEnglish
Title of host publicationICBDC 2018 - Proceedings of 2018 International Conference on Big Data and Computing
PublisherAssociation for Computing Machinery
Pages71-77
Number of pages7
ISBN (Print)9781450364263
DOIs
Publication statusPublished - 28 Apr 2018
Event2018 International Conference on Big Data and Computing, ICBDC 2018 - Shenzhen, China
Duration: 28 Apr 201830 Apr 2018

Publication series

NameACM International Conference Proceeding Series

Conference

Conference2018 International Conference on Big Data and Computing, ICBDC 2018
Country/TerritoryChina
CityShenzhen
Period28/04/1830/04/18

Keywords

  • Category word embedding
  • Climate domain
  • Event detection

Fingerprint

Dive into the research topics of 'Climate event detection algorithm based on climate category word embedding'. Together they form a unique fingerprint.

Cite this