The Construction of a Chinese Collocational Knowledge Resource and Its Application for Second Language Acquisition

3697

Authors

nan

Abstract

AbstractThe appropriate use of collocations is a challenge for second language acquisition. However, high quality and easily accessible Chinese collocation resources are not available for both teachers and students. This paper presents the design and construction of a large scale resource of Chinese collocational knowledge, and a web-based application (OCCA, Online Chinese Collocation Assistant) which offers free and convenient collocation search service to end users. We define and classify collocations based on practical language acquisition needs and utilize a syntax based method to extract nine types of collocations. Totally 37 extraction rules are compiled with word, POS and dependency relation features, 1,750,000 collocations are extracted from a corpus for L2 learning and complementary Wikipedia data, and OCCA is implemented based on these extracted collocations. By comparing OCCA with two traditional collocation dictionaries, we find OCCA has higher entry coverage and collocation quantity, and our method achieves quite low error rate at less than 5%. We also discuss how to apply collocational knowledge to grammatical error detection and demonstrate comparable performance to the best results in 2015 NLP-TEA CGED shared task. The preliminary experiment shows that the collocation knowledge is helpful in detecting all the four types of grammatical errors.

Keywords

nan

Full Content

Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, pages 3254–3263, Osaka, Japan, December 11-17 2016. The Construction of a Chinese Collocational Knowledge Resource and Its Application for Second Language Acquisition Renfen Hu Institute of Chinese Information Pro- cessing, Beijing Normal University 19 Xinjiekouwai Street, Beijing, China irishere@mail.bnu.edu.cn Jiayong Chen and Kuang-hua Chen Department of Library and Information Science, National Taiwan University No.1, Sec. 4, Roosevelt Road, Taipei, Taiwan {d04126001,khchen}@ntu.edu.tw Abstract The appropriate use of collocations is a challenge for second language acquisition. However, high quali- ty and easily accessible Chinese collocation resources are not available for both teachers and students. This paper presents the design and construction of a large scale resource of Chinese collocational knowledge, and a web-based application (OCCA, Online Chinese Collocation Assistant) which offers free and convenient collocation search service to end users. We define and classify collocations based on practical language acquisition needs and utilize a syntax based method to extract nine types of collo- cations. Totally 37 extraction rules are compiled with word, POS and dependency relation features, 1,750,000 collocations are extracted from a corpus for L2 learning and complementary Wikipedia data, and OCCA is implemented based on these extracted collocations. By comparing OCCA with two tradi- tional collocation dictionaries, we find OCCA has higher entry coverage and collocation quantity, and our method achieves quite low error rate at less than 5%. We also discuss how to apply collocational knowledge to grammatical error detection and demonstrate comparable performance to the best results in 2015 NLP-TEA CGED shared task. The preliminary experiment shows that the collocation knowledge is helpful in detecting all the four types of grammatical errors. 1 Introduction Second language (L2) learners often have problems in appropriate use of word collocations (Bahns and Eldaw,1993; Farghal and Obiedat, 1995; Nesselhauf, 2003). For example, English speakers who are learning Chinese are likely to produce incorrect expressions e.g. 漂亮 的 歌 (piaoliang/beautiful de/* ge/song) because “beautiful song” is a reasonable collocation in English. However, 漂 亮 (piaoliang/beautiful) is not an acceptable modifier for 歌 (ge/song) in Chinese (Lu, 1987). Bahns and Eldaw (1993) went deeply into this issue and suggested that leaner's collocational competence does not develop in parallel with general vocabulary knowledge, because words are usually acquired indi- vidually without taking note of their immediate environment (Siyanova and Schmitt, 2008). This problem could be more serious for learning analytic languages such as Chinese, because apart from the negative transfer from the first language (L1), this typical analytic language does not have inflectional morphemes, and uses function words with little lexical meaning to express grammatical relationships. These function words are also highly involved in collocations. The following shows col- location examples with function words: (a) 在 (zai/*) X 上 (shang/on)  on X (b) 诚实 (chengshi/honest) 的(de/*) 人 (ren/person)  honest person (c1) 买 (mai/buy) 东西 (dongxi/something)  buy something (c2) 买 (mai/buy) 了(le/*) 东西 (dongxi/something)  have bought something (c3) 买 (mai/buy) 着 (zhe/*) 东西 (dongxi/something)  be buying something Example …

Content truncated to 500 words

Labeled Text

AbstractThe appropriate use of collocations is a challenge for second language acquisition. However, high quality and easily accessible Chinese collocation resources are not available for both teachers and students. This paper presents the design and construction of a large scale resource of Chinese collocational knowledge, and a web-based application (OCCA, Online Chinese Collocation Assistant) which offers free and convenient collocation search service to end users. We define and classify collocations based on practical language acquisition needs and utilize a syntax based method to extract nine types of collocations. Totally 37 extraction rules are compiled with word, POS and dependency relation features, 1,750,000 collocations are extracted from a corpus for L2 learning and complementary Wikipedia data, and OCCA is implemented based on these extracted collocations. By comparing OCCA with two traditional collocation dictionaries, we find OCCA has higher entry coverage and collocation quantity, and our method achieves quite low error rate at less than 5%. We also discuss how to apply collocational knowledge to grammatical error detection and demonstrate comparable performance to the best results in 2015 NLP-TEA CGED shared task. The preliminary experiment shows that the collocation knowledge is helpful in detecting all the four types of grammatical errors.

Publication Details

Paper ID:
3697
Year:
2016
Publication:
TIPSTER
Publication Type:
conf
DOI:
nan