Honorific agreement and plural copying revisited: Experimental and deep learning approach
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
The study investigated two types of subjecthood diagnostics in Korean using two different kinds of approaches. Based on the previous studies on subjecthood diagnostics in Korean, this paper examined the two subjecthood diagnostics: Honorific Agreement (HA) and Plural Copying (PC). For the experimental analysis, this study adopted the analysis results of Kim et al. (2017). For the deep learning analysis, this paper employed the KR-BERT for the deep learning model and three sources of data sets: the Korean version of the Corpus of Linguistic Acceptability (K-CoLA), the Sejong Morphologically-Analyzed Corpus, and the extended target sentences of Kim et al. (2017). Two separate experiments were conducted in the deep learning analysis. In the first experiment, the KR-BERT was trained only with the K-CoLA, and the target sentences were analyzed. In the second experiment, the KR-BERT was trained with the K-CoLA and the sentences from the Sejong corpus, and the acceptability scores of the target sentences were measured. The acceptability scores were measured with the numeric scores using the algorithms in Lee (2021). After the experiments with the deep learning models, the scores were normalized and were statistically analyzed with Generalized Linear Models (GLMs). Through the two experiments, the following fact was observed: both HA and PC did not show similar tendencies of experimental results with human participants in the first experiment, but they did in the second experiment. The analysis results demonstrated that both HA and PC could be used as subjecthood diagnostics but that they played significant roles only when native speakers were exposed to enough examples.
Abstract
The study investigated two types of subjecthood diagnostics in Korean using two different kinds of approaches. Based on the previous studies on subjecthood diagnostics in Korean, this paper examined the two subjecthood diagnostics: Honorific Agreement (HA) and Plural Copying (PC). For the experimental analysis, this study adopted the analysis results of Kim et al. (2017). For the deep learning analysis, this paper employed the KR-BERT for the deep learning model and three sources of data sets: the Korean version of the Corpus of Linguistic Acceptability (K-CoLA), the Sejong Morphologically-Analyzed Corpus, and the extended target sentences of Kim et al. (2017). Two separate experiments were conducted in the deep learning analysis. In the first experiment, the KR-BERT was trained only with the K-CoLA, and the target sentences were analyzed. In the second experiment, the KR-BERT was trained with the K-CoLA and the sentences from the Sejong corpus, and the acceptability scores of the target sentences were measured. The acceptability scores were measured with the numeric scores using the algorithms in Lee (2021). After the experiments with the deep learning models, the scores were normalized and were statistically analyzed with Generalized Linear Models (GLMs). Through the two experiments, the following fact was observed: both HA and PC did not show similar tendencies of experimental results with human participants in the first experiment, but they did in the second experiment. The analysis results demonstrated that both HA and PC could be used as subjecthood diagnostics but that they played significant roles only when native speakers were exposed to enough examples.
참고문헌 (55)
- Bard, Ellen, "Magnitude estimation of linguistic acceptability", Language, 72(1), pp.32~68 (1996)
- 최재웅, "Obligatory Honorification and the Honorific Feature", 생성문법연구, 14(4), pp.545~559 (2004), 한국생성문법학회
- Cowart, Wayne, "Experimental syntax: Applying objective methods to sentence judgments" (1997), Sage Publications
- Devlin, Jacob, "BERT: Pre-training of deep bidirectional transformers for language understanding", arXiv preprint arXiv: 1810.04805, pp.04805~ (2019)
- Falk, Yehuda, "Subjects and universal grammar" (2006), Cambridge University Press
- Gravetter, Fedrick, "Statistics for the behavioral sciences" (2013), Wadsorth
- Gries, Stefan Th, "Statistics for linguistics with R : A practical introduction" (2013), Gruyter
- Han, Eunjoo, "Honorification in Korean", pp.1990
- Heyock, Caroline, "Syntactic predication in Japanese", Journal of East Asian Linguistics, 2(2), pp.167~211 (1993)
- Hong, Ki-Sun, "Argument selection and Case-marking in Korean", PhD, pp.1991
- Hong, Ki-Sun, "Subjecthood tests in Korean", Language Research, 30(1), pp.99~136 (1994)
- Hong, Ki-Sun, "Hankwuke-uy kyekcwungchwul kwumwun-kwa tamhwakwuco (Multiple case construction in Korean and the discourse structure)", Proceedings of 2014 Korean Society for Language and Information (KSLI) Annual Conference, pp.101~109 (2014)
- Kang, Beom-Mo, "Pemcwu mwunpep: Hankwuke-uy hyengthaylon, thongsalon, thaipnonlicek uymilon (Categorial grammar: The morphology, syntax, and type-logical semantics of Korean)" (2002), Korea University Press
- Keller, Frank, "Gradient in grammar : Experimental and computational aspects of degrees of grammaticality", PhD, pp.2000
- 김정석, "A Note on Subject Honorification in Korean", 생성문법연구, 32(1), pp.153~167 (2022), 한국생성문법학회[DOI]
- 김지혜, "Obligatory Control and Coordinated Deletion as Korean Subject Diagnostics: An Experimental Approach", 언어와 정보, 19(1), pp.75~101 (2015), 한국언어정보학회
- 김지혜, "Honorific Agreement and Plural Copying as Korean Subjecthood Diagnostics: An Experimental Approach", 현대문법연구, (93), pp.119~144 (2017), 현대문법학회[DOI]
- Kim, Jong-Bok, "Korean honorification : A kind of expressive meaning", Journal of East Asian Linguistics, 16(4), pp.303~336 (2007)
- Kuh, Hakan, "Harvard studies in Korean linguistics II", pp.239~250 (1987), Department of Linguistics, Harvard University
- Lee, Chungmin, "Grammatical constraints on honorific expressions in Korean", pp.1990
- Lee, Hyo-Sang, "Causatives in Korean and the binding hierarchy", pp.138~153 (1985), University of Chicago Press
- Lee, Ik-Hwan, "Harvard Studies in Korean Linguistics VII", pp.287~296 (1997), Department of Linguistics, Harvard University
- Lee, Junbum, "KcBERT: Korean comments BERT", Proceedings of the 32nd Annual Conference on Human & Cognitive Language Technology, pp.437~440 (2020)
- Lee, Sangah, "KR-BERT: A small-scale Korean-specific language model", arXiv preprint arXiv:2008.03979 (2020)
- Lee, Yong-hun, "Corpus linguistics and statistics with R" (2016), Hankook Publishing Co
- 이용훈, "English Island Constraints Revisited: Experimental vs. Deep Learning Approach", 영어학연구, 27(3), pp.21~45 (2021), 한국영어학학회[DOI]
- 이용훈, "Lexical Effects in Island Constraints: A Deep Learning Approach", 언어학, 30(1), pp.179~201 (2022), 대한언어학회[DOI]
- 이용훈, "Reflexive Binding and Adjunct Control as Subject Diagnostics in Korean: An Experimental Approach", 언어연구, 31(2), pp.427~449 (2015), 한국현대언어학회[DOI]
- 이용훈, "A multi-level analysis of subjecthood diagnostics in Korean", 언어연구, 32(3), pp.671~691 (2015), 언어정보연구소[DOI]
- McCoy, Thomas, "Does syntax need to grow on trees? Sources of hierarchical inductive bias in sequence-to-sequence networks", Transactions of the Association for Computational Linguistics, 8, pp.125~140 (2020)
- The final report of 21th century Sejong project, "National Institute of Korean Language", National Institute of Korean Language, ~ (2007)
- Park, Byung-Soo, "On the multiple subject constructions in Korean", Linguistics, 100, pp.63~76 (1973)
- Park, Ki-Seong, "The semantics and pragmatics of Case marking in Korean: A role and reference grammar account", PhD, pp.1995
- 박권식, "최소대립 문장쌍을 활용한 한국어 사전학습모델의 통사 연구 활용 가능성 검증", 언어와 정보, 25(3), pp.1~21 (2021), 한국언어정보학회
- Park, Sungjoon, "KLUE: Korean language understanding evaluation", arXiv preprint arXiv:2105.09680, pp.09680~ (2021)
- Pires, Telmo, "How multilingual is multilingual BERT?", arXiv preprint arXiv:1906.01502 (2019)
- R Core Team, "R : A language and environment for statistical computing" (2022), R Foundation for Statistical Computing
- Schütze, Carson, "The empirical base of linguistics : grammaticality judgments and linguistic methodology" (1996), University of Chicago Press
- Schütze, Carson, "On Korean ‘Case stacking’ : The varied functions of the particles-ka and-lul", The Linguistic Review, 18, pp.193~232 (2001)
- Sennrich, Rico, "Neural machine translation of rare words with subword units", arXiv preprint arXiv:1508.07909, pp.07909~ (2016)
- Song, Sanghoun, "An empirical study of honorific mismatches in Korean", Language Sciences, 75, pp.47~71 (2019)
- Song, Suk-Choong, "Bare Plural Marking and Ubiquitous Plural Marker in Korean", Proceedings of the 11th regional meeting of Chicago Linguistic Society, pp.536~546 (1975), University of Chicago Press
- Suh, Cheong-Soo, "Papers in Korean Linguistics", pp.297~304 (1977), Hornbeam Press
- Wang, Alex, "GLUE: A multi-task benchmark and analysis platform for natural language understanding", arXiv preprint arXiv: 1804.07461 (2019)
- Warstadt, Alex, "Neural network acceptability Judgments", arXiv preprint arXiv:1805.12471 (2019)
- Wilcox, Ethan, "What syntactic structures block dependencies in RNN language models?", arXiv preprint arXiv:1905.10431 (2019)
- Williams, Edwin, "Predication", Linguistic Inquiry, 11(1), pp.203~238 (1980)
- Yoon, James, "Some queries concerning the syntax of multiple subject constructions in Korean", Studies in the Linguistic Sciences, 16(3), pp.215~236 (1986)
- Yoon, James, "Harvard Studies in Korean Linguistics II", pp.138~162 (1987), Department of Linguistics, Harvard University
- Yoon, James, "Non-nominative subjects 2", pp.265~314 (2004), Mouton
- Yoon, James, "Subjecthood and subject properties in multiple subject constructions", Handout provided in East Asian Linguistics Seminar (2008), Oxford University
- Yoon, James, "The distribution of subject properties in multiple subject constructions", Japanese, 2009, pp.19, 64~83
- Yoon, James, "The handbook of Korean linguistics", pp.79~97 (2015), John Wiley &Sons, Inc
- Yoon, Jong-yurl, "Harvard Studies in Korean Linguistics III", pp.383~394 (1989), Department of Linguistics, Harvard University
- Youn, Cheong, "A relational analysis of Korean multiple nominative constructions", PhD, pp.1990