2016, Vol., No.48

The Japanese Language Association of Korea 2023 KCI Impact Factor : 1.0

pISSN : 1229-7275
https://journal.kci.go.kr/jlak

This paper explains the preliminary research of an Automatic Scoring System being developed through which Japanese learners can submit compositions for pre-set tasks to an online scoring application. We investigated the ‘YNU written language corpus’ to find a method by which completion of tasks can be automatically determined. The aims of this investigation were to determine the feasibility of automatic content evaluation and appropriate tasks for this application. We extracted nouns from 30 compositions from each of 6 tasks written by native Korean speakers(NKS)、 and the same number of compositions written by native Japanese speakers(NJS)、 using the morpheme analyzer “Cha-Sen”. We then categorized the extracted “mandatory words”(MW)using the 『Bunruigoihyo』. We also compared the number of MW with the total number of nouns used for each task、 and found certain tasks were more appropriate for automatic content analysis、 because the number of MW was small. We found differences in both the number and types of MW used by NKS versus those used by NJS. We also discovered that while NJS did not use unnecessary nouns to complete the tasks、 NKS did. Our analysis of the participants' submissions using the『Bunruigoihyo』showed that NJS used a wider variety of synonyms to express the same notions than NKS. Based on the results of this investigation、 we came to the conclusion that an automatic scoring system that focuses on MW can not only evaluate the content of compositions、 but also determine the fulfillment of pre-set tasks.

The purpose of this study is to quantitatively analyze lower Japanese nonnative speakers’ conversations at Palau. We analyzed the utterances by comparing Japanese native speakers and Japanese learners for similarities or differences. As a result, he(TN) used "back-channel feedback" and "discourse-marker" as much as these speakers. Needless to say, he could communicate more smoothly in Japanese. In addition, this study found "Short turn-taking" by "back-channel feedback" and "discourse-marker". On the other hand, there were non-typical expressions of "back-channel feedback おお", function of " repetition " and so on. As the significance of this study, we provided limited data about lower Japanese nonnative speakers’ conversations at Palau, and such empirical analyses of actual data will enable us to offer effective and realistic materials for instructing non-native speakers in the techniques of Japanese conversation construction.

The aim of this paper is to examine the phenomenon of nouns that express a spatial meaning when combined with motion verbs, even though they normally do not convey that meaning.Furthermore, this paper examines how the degree of the shift to the spatial meaning differs between different nouns and the reasons for that difference. In order to examine these phenomena, I conducted a survey among Japanese native speakers, using a questionnaire that contained sentences with motion expressions. The results of the survey indicate that the degree of acceptability is higher with some nouns than with others, regardless of the verbs. However, since in the case of iku ‘go’, tsuku ‘arrive’ and mukau ‘move toward’, the degree of acceptability varies depending on the nouns they are combined with, it can be said that the type of verb also affects acceptability. Furthermore, the results of the survey show that the reasons for differences in the degrees of acceptability are closely related to semantic characteristics of motion verbs, the degree of motion, and size of the physical objects of nouns.