| Budget Amount *help |
¥1,700,000 (Direct Cost: ¥1,700,000)
Fiscal Year 2005: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2004: ¥800,000 (Direct Cost: ¥800,000)
|
|---|
| Research Abstract |
This study aims to suggest a guidance for lexical teaching of Korean as a foreign language, by contrasting the collocational structures of Korean and Japanese language.
The results are as follows.
1. Building a tagged parallel corpus of Korean and Japanese
A parallel corpus of Korean-Japanese newspaper articles, respectively composed of about 500,000words, is morphologically tagged by IKMA (Intelligent Korean Morphological Analyzer) and ChaSen (for Japanese).
2. Extracting Korean collocation structures
For the words that have frequency more than 100, the collocates within the span +,-5 are extracted by using 'WordParser'. Using t-score and relative frequency, the collocations in Korean corpus are identified. To supplement the small-corpus-weakness, the lexical collocations listed by previous studies are identified in parallel corpus.
3. Grouping Korean collocation structures
Korean collocations are classified by the number of components, as bigram and multigram collocations. From the viewpoint of lexicon and grammar, Korean collocations show 3 categories of grammatical collocation, lexical collocation, and categorial collocation.
4. Contrasting Korean collocations to their Japanese equivalencies
Concerning the correspondence between Korean collocations and their Japanese equivalencies, there are 3 modes of correspondency.
5. Applying collocational information to TKFL (Teaching Korean as a Foreign Language)
Collocational information is turned out to be available for teaching synonyms, onomatopoeia, words without Japanese partners, and word order. Finally, a revised collocation list for KTFL is suggested.
|
