Analyzing Subjecthood Tests in Korean Using a Cross-Validation

다국어 초록 (Multilingual Abstract)

Many diagnostics have been proposed to examine subjecthood in Korean, some of which were investigated in terms of their validity as subject diagnostics through a series of experimental studies. This paper takes the experimental data from three experimental studies (Kim et al., 2015; Lee et al., 2015a; Kim et al., 2017) that were previously conducted and examines the validity of six diagnostic tests using a cross-validation and a machine learning method. The experimental data were evenly divided into 10 parts (10-fold cross-validation): Nine of them were used for machine learning (training data set) and the remaining one was for the testing (test data set). The analysis was conducted 10 times, and the average values were taken. A simple linear regression was conducted for machine learning. Through the analysis, the following results were observed: (i) The accuracy of the diagnostics for SSC was much higher than that of MSC (SSC: 81,78, MSC: 68.66), (ii) Adjunct Control was the most accurate for SSC, whereas Coordinated Deletion for MSC, (iii) HA and PL had low performance. The results also seem to imply that there are some parts which cannot be explained by a simple linear model but can be analyzed with non-linear modeling of language.

참고문헌 (Reference)

1 Park, K.-S., "The semantics and pragmatics of case marking in Korean: A role and reference grammar account" State University of New York 1995

2 Falk, Y., "Subjects and Universal Grammar" Cambridge University Press 2006

3 Heycock, C., "Subjects and Predication in Korean and Japanese" 25 (25): 755-792, 1989

4 Yoon, J. H. -S., "Subjecthood and Subject Properties in Multiple Subject Constructions" Oxford University 2008

5 Hong, K.-S., "Subjecthood Tests in Korean" 30 : 99-136, 1994

6 Gravetter, F., "Statistics for the Behavioral Sciences" Wadsworth 2013

7 Gries, S. Th, "Statistics for Linguistics with R: A Practical Introduction" Guyter 2013

8 Samuel, A, "Some Studies in Machine Learning Using the Game of Checkers" 3 (3): 210-229, 1959

9 Yoon, J. H. -S, "Some Queries Concerning the Syntax of Multiple Subject Constructions in Korean" Department of Linguistics, University of Illinois 215-236, 1986

10 이용훈, "Reflexive Binding and Adjunct Control as Subject Diagnostics in Korean: An Experimental Approach" 한국현대언어학회 31 (31): 427-449, 2015

11 Yoon, J. H. -S, "Raising of Major Arguments in Korean and Japanese" 25 : 615-653, 2007

12 Geisser, S., "Predictive Inference" Chapman and Hall 1993

13 이용훈, "Predicting the Occurrence of the English Modals Can and May Using Deep Neural Networks" 현대문법학회 (96) : 167-189, 2017

14 Kang, B. -M., "Pemcwu Mwunpep: Hankwuke-uy Hyengthaylon, Thongsalon, Thaipnonlicek Uymilon (Categorial Grammar: The Morphology, Syntax, and Type-Logical Semantics of Korean)" Korea University Press 2002

15 Park, B. -S, "On the Multiple Subject Constructions in Korean" 100 : 63-76, 1973

16 Schütze, C, "On Korean ‘Case Stacking’: The Varied Functions of the Particles -ka and -lul" 18 : 193-232, 2001

17 김지혜, "Obligatory Control and Coordinated Deletion as Korean Subject Diagnostics: An Experimental Approach" 한국언어정보학회 19 (19): 75-101, 2015

18 Yoon, J. H.-S., "Non-nominative Subjects, volume 2" Mouton de Gruyter 265-314, 2004

19 Mitchell, T., "Machine Learning" McGraw Hill 1997

20 Andrews, A., "Language Typology and Syntactic Description" Cambridge University Press 132-223, 1985

21 김지혜, "Honorific Agreement and Plural Copying as Korean Subjecthood Diagnostics: An Experimental Approach" 현대문법학회 (93) : 119-144, 2017

22 Yoon, J. -Y, "Harvard Studies in Korean Linguistics III" 383-394, 1989

23 Lee, I. -H, "Harvard Studies in Korean Linguistics II" 287-296, 1987

24 Teng, S-H, "Double Nominatives in Chinese" 50 : 455-473, 1974

25 Hong, K. -S., "Argument Selection and Case-Marking in Korean" Stanford University 1991

26 Kim, J.-H., "An Experimental Study of Subject Properties in Korean Multiple Subject Constructions (MSCs)" Kyung Hee University 183-190, 2016

27 이용훈, "A multi-level analysis of subjecthood diagnostics in Korean" 언어정보연구소 32 (32): 671-691, 2015

28 Kohavi, R., "A Study of Cross-validation and Bootstrap for Accuracy Estimation and Model Selection" Morgan Kaufmann 2 (2): 1137-1143, 1995

29 Youn, C., "A Relational Analysis of Korean Multiple Nominative Constructions" State University of New York 1990