2006 Fiscal Year Final Research Report Summary
Application of the 60,000 Kanji Database over the Internet for overseas researchers of the Japanese language.
| Project/Area Number |
16520290
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Japanese linguistics
|
|---|
| Research Institution | The National Institute for Japanese Language |
|---|
Principal Investigator |
YOKOYAMA Shoichi The National Institute for Japanese Language, Department of Language Research, Leader, 研究開発部門, グループ長 (60182713)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
YANAGISAWA Yoshiaki The National Institute for Japanese Language, Center for Research in Japanese as a Second Language, Director, 日本語教育情報基盤センター, センター長 (80249911)
YONEDA Junko The National Institute for Japanese Language, Department of Language Research, Researcher, 研究開発部門, 研究員 (60311156)
|
|---|
| Project Period (FY) |
2004 – 2006
|
| Keywords | Japanese-language resources / kanji dictionary / preference / familiarity / logistic regression analysis / generalized matching law / Internet / kanji variant |
|---|
| Research Abstract |
This research investigated basic phenomena, which are involved in digital transactions of data on Japanese kanji characters over the Internet. It was our motivation to help overseas researchers of the Japanese language to access data on kanji via commonly-used web browsers on computers without Japanese encoding systems. The research specifically focused on the graphic attributes of kanji based on the 60,000 Kanji Database for electronic government.
A major purpose of the research was to investigate the application and impacts of the character glyphs of the 60,000 Kanji Database, which were selected as a representative standard of kanji, for the purpose of the research, due to the following two reasons : (1) technical reliability; and (2) academic quality. The character glyphs of the database were designed for reliable presentation of characters' fonts and shapes across technical environments, and thus, they were ideal for our target users in diverse technical environments. Furthermore,
… More
the database was developed collaboratively by the Japanese Standards Association and the National Institute for Japanese Language, exhibiting accurate and profound information based on solid academic framework. Such a qualitative aspect of the database satisfies the demand of our target users.
The research collected empirical evidence of familiarity with and preference for these characters glyphs, which were analyzed by multivariate analysis methods, including logistic regression analysis. The analyses revealed that the preference for character glyphs can be accurately estimated based on frequency data obtained from existing corpora. The results also indicated that such preference behavior is well accounted for by a statistical model based on the generalized matching law. These findings indicated that linguistic data, such as corpora, may estimate psychological behavior, such as preference, suggesting that long-term impacts of standards may be predicted based on existing data, though our findings leaves room for further investigation to validate the predictive power of the model as well as existing data. Less
|
