This paper aims to review diachronic processes of Korean medical terminology in terms of the refining of words, and to discuss some word formation issues about Korean medical terminology in the dictionary of Urimalsaem. In medical terminology, there have been steady attempts to simplify the terminology since the late 1980's. The Anatomical Terminology (3rd edition, 1990) is a representative dictionary which replaced many Sino-Korean words originating from Japan with native Korean expressions. The Medical Terminology (3rd edition, 1992) adopted vocabulary of the Anatomical Terminology, and many native Korean medical terms appeared in the fourth edition (2001) of the Medical Terminology. There are some characteristics of word formation in the new medical terminology. Overall the move from Sino-Korean words to native Korean words, derivation using affixes, the active use of verbs, word formation based on syntactic constructions, the use of bound root, and characteristic combinations by shape features are mentioned.

In this paper, I critically review some descriptions of the morphological terminology in the Standard Korean Dictionary (henceforth, SKD). The findings are as follows. First, there are 3,480 linguistics terms and 225 morphology terms in the SKD, and most of the morphology terms are the expressions of word-formation. Second, the definitions of some morphology terms in the SKD(English equivalents shown), such as variant, allomorph, nominal compound, verbal compound, adjectival compound, acronym, collocation, blend, conjugation form, differ from those of descriptive grammar. Third, some words not listed in the SKD are presented as the definition and the immediate constituents of the headword. Fourth, some related morphology terms in the SKD, such as simple word, complex word, immediate constituent, are not presented consistently.

This paper studies the meaning and function of the two spoken language words, eccay and eccenci. Eccihata is an interrogative with the meaning of asking the reason and the manner. However, the inflected and diminished forms, eccay and eccenci, carry various senses other than the meanings of the original, uninflected form. Dictionaries do not incorporate language change or the spoken language, and so the meanings used in the real life often do not match the meanings found in the dictionary. Though the first straightforward way to approach a foreign language is to use a dictionary, there is a limit, depending on the dictionary, in learning the functions and meanings actually used in the spoken language. This paper studies the meanings and functions of the two dimished forms as used in real life by the speakers of Korean. Further research is called for regarding the compilation of dictionaries and study books.

The purpose of this study is to examine the variables which could potentially contribute to vocabulary inference by learners of Korean. A total of 215 learners of Korean have been invited and among them 203 students were assessed in a questionnaire survey on vocabulary reasoning strategy, vocabulary learning attitude and vocabulary exposure experience. The results are as follows. First, according to factor analysis, three factors are found: vocabulary reasoning strategy, vocabulary learning attitude, and vocabulary exposure experience. In addition, based on structural equation modeling (SEM), vocabulary reasoning strategy and vocabulary learning attitude have a significant direct effect on vocabulary inference. Second, all of the three factors are positively correlated with one another. Third, vocabulary reasoning strategy and vocabulary learning attitude are strong predicators of vocabulary inference results.

The aim of this study is to find out how the colloquial written language reflects the oral features of the corresponding spoken language. For this purpose, phonological reduction found in messenger corpora has been analyzed in phoneme and syllable units and compared with that found in spoken language using a quantitative method. Some words show a similar reduction pattern as spoken language patterns. However, it is also observed that the weakening applies selectively or in a different direction than in spoken language. An unfamiliar form that does not appear in spoken language is also observed. To sum up, phonological reduction in messenger corpora is partly similar to that in spoken language, but it also has an independent aspect that is difficult to incorporate into the orality framework. This result implies that colloquial written language should be considered as an independent register.

This discussion is an analysis of the incomplete version of the study book, A Language Book of Korean, written by Allen, a foreign missionary who entered Korea during the Joseon era for medical missionary purposes. This book is worthy as a source for how early Joseon language textbooks were compiled, as efforts to learn Korean were revealed for the first time. In this book, we examine the characteristics of vocabulary items by dividing them into general vocabulary and medical terminology sections. In the general vocabulary section, vocabulary used in everyday life is introduced to enable communication in Joseon. The medical terminology section introduces Korean medical terms for medical missionary purposes. In the latter case, the words are presented in alphabetical order (i.e., A to Z) for the convenience of foreign missionaries such as the author himself.

Machine translation commonly involves both analysis and generation across different human languages. This implies that parallel corpora of a large size are essential to create a theoretically reliable and practically robust translation model. However, as is well known, the parallel corpora between Korean and English are insufficient. In this respect, this study expands the data by means of English Resource Grammar (Flickinger 2000) to improve the translation model between the languages. Then, it looks at whether the neural machine translation model performs better with the augmented data. Unfortunately, it turns out the translation models based on augmented data exhibit rather lower BLEU scores. This study further discusses the reason for the unsatisfactory scores and raises the necessity of human evaluation as a next step.