규범과 현실 사이의 거리 좁히기: 국어 어휘 빈도 조사 방법론의 정립을 위한 시론

Bridging the Gap Between the Norm and the Practice: A tentative for setting up word frequency analysis methodology in Korean

In analysing word frequencies of modern Korean written text, the most troublesome problem is how to treat compound words and spacing disorders. In this paper, author have propsed a tentative method to fix the problem. This can be summurized as follows. (1) It is preferable to include multi-word lexical units in word frequency analysis, because there are so many linguistic elements which average Koreans can not determine whether it is a word or a systactic structure. Among them, phrasal verbs which have the structure of 'noun+ (particle)+verb', noun and particle combinations that are on the stage of lexicalization are wellknown. (2) To solve spaceing disorder problem, the steps of word frequency analysis shoul be devided (1) Analysing the frequencies of every tokens in the whole corpus. (2) Extracting the frequencies of each lexical unit. which can be treated as a word by using the results of (1). (3) Using the list of the types, check off every linguistic type which made of more than one lexical elements, these are lexical units which may show spacing-disorders. (4) Using KWIC (keyword in context) and collocation extracting tool. check off every collocates of each simple word which compose compounds or phrasal words. (5) Settle the results of (2) by the results of (4). (6) List the frequency hierachically, based on word-family strucure of each word-base.