CONFERENCE PROCEEDINGS
Advanced Search
Home > Proceedings > Volume 12250 > Article
Paper
19 May 2022 Combining error oversampling and multi-task learning for Chinese meteorological alert text correction
Yu Mei, Wei Tang, Muhua Wang, Jianzhong Hui, Hanhua Qu, Kuoyin Wang
Author Affiliations +
Proceedings Volume 12250, International Symposium on Computer Applications and Information Systems (ISCAIS 2022); 122500P (2022) https://doi.org/10.1117/12.2639645
Event: International Symposium on Computer Applications and Information Systems (ISCAIS2022), 2022, Shenzhen, China
Abstract
Chinese meteorological alert text issued by meteorological authorities needs to be free of spelling errors. Automatic spelling error correction can find text errors and give correction suggestion. Most existing studies focus on open domain text correction such as news etc., however, the methods for vertical domain text correction such as meteorological alert text has not been well studied. In this work, we utilize the template feature of meteorological alert text and propose error oversampling strategy to enhance the correction model training. As for the correction model, we use multi-task learning to train the correction model by accounting error detection and error correction simultaneously. Experimental results on real-world alert texts show that our proposed method is significantly better than the baseline, exceeding the baseline by 3% on F1 measure.
© (2022) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Citation Download Citation
Yu Mei, Wei Tang, Muhua Wang, Jianzhong Hui, Hanhua Qu, and Kuoyin Wang "Combining error oversampling and multi-task learning for Chinese meteorological alert text correction", Proc. SPIE 12250, International Symposium on Computer Applications and Information Systems (ISCAIS 2022), 122500P (19 May 2022); https://doi.org/10.1117/12.2639645
ACCESS THE FULL ARTICLE
ORGANIZATIONAL
Sign in with credentials provided by your organization.
INSTITUTIONAL
Select your institution to access the SPIE Digital Library.
SELECT YOUR INSTITUTION
PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
PERSONAL SIGN IN
No SPIE Account? Create one
PURCHASE THIS CONTENT
SUBSCRIBE TO DIGITAL LIBRARY
50 downloads per 1-year subscription
Members: $195
Non-members: $335 ADD TO CART
25 downloads per 1 - year subscription
Members: $145
Non-members: $250 ADD TO CART
PURCHASE SINGLE ARTICLE
Includes PDF, HTML & Video, when available
Members: $17.00
Non-members: $21.00 ADD TO CART
PROCEEDINGS
 7 PAGES
DOWNLOAD PAPER SAVE TO MY LIBRARY
GET CITATION
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Meteorology
Computer programming
Data corrections
Machine learning
RELATED CONTENT
Subscribe to Digital Library
Receive Erratum Email Alert
Back to Top