Correcting Chinese Word Usage Errors for Learning Chinese as a Second Language

Proceedings of the 27th International Conference on Computational Linguistics, pages 2410–2422 Santa Fe, New Mexico, USA, August 20-26, 2018. 2410 Correcting Chinese Word Usage Errors for Learning Chinese as a Second Language Yow-Ting Shiue1, Hen-Hsen Huang1, and Hsin-Hsi Chen1,2 1Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan 2MOST Joint Research Center for AI Technology and All Vista Healthcare, Taipei, Taiwan orina1123@gmail.com, hhhuang@nlg.csie.ntu.edu.tw, hhchen@ntu.edu.tw Abstract With more and more people around the world learning Chinese as a second language, the need of Chinese error correction tools is increasing. In the HSK dynamic composition corpus, word usage error (WUE) is the most common error type. In this paper, we build a neural network model that considers both target erroneous token and context to generate a correction vector and compare it against a candidate vocabulary to propose suitable corrections. To deal with potential alternative corrections, the top five proposed candidates are judged by native Chinese speakers. For more than 91% of the cases, our system can propose at least one acceptable correction within a list of five candidates. To the best of our knowledge, this is the first research addressing general-type Chinese WUE correction. Our system can help non-native Chinese learners revise their sentences by themselves. Title and Abstract in Chinese 非中文母語學習者中文寫作用詞錯誤之更正 以中文為第二語言的學習者與日俱增，遍及全球，對於中文更正工具的需求也相應而 生。在HSK動態作文語料庫中，用詞錯誤（WUE）是中文學習者最常犯的錯誤類型。在 這篇論文中，針對中文用詞的更正，我們建立了專屬的神經網路模型，同時參酌目標錯 誤詞彙及其前後文資訊，以產生推薦的更正方式。經過中文母語人士的評估，在91%的 測試案例中，系統所推薦的前五名候選詞彙，至少有一個是適當的更正方式。據我們所 知，這是目前第一個全面性的中文用詞錯誤更正之研究。我們的系統可以幫助非中文母 語人士自主修正其所寫的中文句子，促進華語文教學之成效。 1 Introduction Grammatical error correction (GEC) tools can help language learners revise their writing. Chinese GEC tools are in high demand since Chinese has become an increasingly popular second language worldwide. Despite the increasing need, most of the existing studies on GEC are based on English learner data. The method of correcting sentences in Chinese, a language which differs substantially from English in major aspects such as the morphological structure and the distribution of learner errors, has not yet been fully developed. This paper focuses on the correction of Chinese word usage errors (WUEs). According to the definition of Shiue and Chen (2016), a WUE refers to an incorrect token that involves morphological, syntactical, or semantical problems. The token is either an incorrect word form, or a correct existent word that is improper for its context. Given a token in a sentence segment that is known to be erroneous, we aim to generate a suitable correction for it. The criteria for a suitable correction are: • Correctness: After substituting the erroneous token with the correction token, the result is a syn- tactically and semantically correct Chinese sentence segment. This work is licensed under a Creative Commons Attribution 4.0 International License. License details: http:// creativecommons.org/licenses/by/4.0/ 2411 • Similarity: The meaning of the correction is close to the writer’s intended meaning. We discuss the criteria with the following example sentence segments. (E1-1) *生活方式已經猛 猛 猛烈 烈 (E1-2) *生活方式已經暴烈地改變了 ( The way of living has been fiercely changed. ) 暴 暴烈 烈 (E1-3) 生活方式已經緩烈地改變了 ( ... has been overpoweringly changed. ) 緩 緩慢 慢 (E1-4) 生活方式已經劇慢地改變了 ( ... has been slowly changed. ) 劇 劇烈 烈 烈地改變了 ( ... has been dramatically changed. …

Content truncated to 500 words