1995 Fiscal Year Final Research Report Summary
Organization of English Historical Corpus for Linguistic Research
| Project/Area Number |
06610425
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
英語・英米文学
|
|---|
| Research Institution | HOKKAIDO UNIVERSITY |
|---|
Principal Investigator |
SONODA Katsuhide Hokkaido Univ., Inst. of Language and Culture Studies, Assoc. Prof., 言語文化部, 教授 (70113694)
|
|---|
| Project Period (FY) |
1994 – 1995
|
| Keywords | History of English / Middle English / Corpus / Tagging / Parsing |
|---|
| Research Abstract |
I have designed a tag set for Middle English texts with current theoretical developments of linguistics in mind. The tag set annotates texts in order to fascilliate computational processing of the texts for linguistic research. It comprises symbols for word-tagging and syntactic-parsing. The tags for showing document structures are not included, for which purposes the COCOA format or SGML are already available.
An experimental corpus is prepared based on Margaret Paston's letters. The corpus of 68,000 words has been heavily tagged with the set .
Using the corpus, I conducted pilot studies on spelling, prepositions, negatives, and word order . The tag set has proved to be useful for linguistic analyzes.
Some experiments were conducted using morphological analysis software called PC-KIMMO.It has turned out that the software is not so powerful as to make automatic tagging of Middle English texts feasible.
|
