Automatic Arabic Grammatical Error Correction based on Expectation-Maximization routing and target-bidirectional agreement
作者:
Highlights:
• We proposed an automatic Arabic grammar error correction.
• we proposed a semi-supervision noising method to construct synthetic parallel training data.
• We proposed to use dynamic linear combination and EM routing algorithm with seq2seq Transformer.
• We introduced a regularization bidirectional agreement method to overcome the exposure bias problem.
摘要
•We proposed an automatic Arabic grammar error correction.•we proposed a semi-supervision noising method to construct synthetic parallel training data.•We proposed to use dynamic linear combination and EM routing algorithm with seq2seq Transformer.•We introduced a regularization bidirectional agreement method to overcome the exposure bias problem.
论文关键词:Automatic grammar error correction,Capsule network,Expectation-maximization routing
论文评审过程:Received 24 August 2021, Revised 11 November 2021, Accepted 6 January 2022, Available online 13 January 2022, Version of Record 2 February 2022.
论文官网地址:https://doi.org/10.1016/j.knosys.2022.108180