Automatic Error Detection in Spanish Texts Written byJapanese Learners
| Project/Area Number |
22820047
|
|---|
| Research Category |
Grant-in-Aid for Research Activity Start-up
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Foreign language education
|
|---|
| Research Institution | Aichi Prefectural University |
|---|
Principal Investigator |
VALVERDE Pilar 愛知県立大学, 外国語学部, 講師 (10588205)
|
|---|
| Project Period (FY) |
2010 – 2012
|
| Project Status |
Completed (Fiscal Year 2012)
|
| Budget Amount *help |
¥3,373,578 (Direct Cost: ¥2,595,060、Indirect Cost: ¥778,518)
Fiscal Year 2012: ¥1,033,578 (Direct Cost: ¥795,060、Indirect Cost: ¥238,518)
Fiscal Year 2011: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2010: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
|
|---|
| Keywords | スペイン語 / 言語習得 / コーパス / 誤り検出 / 制約文法 / 学習者 / エラー / エラー自動抽出 / 難易構文 / 第二言語習得 / 誤り自動検出 / 学習者コーパス / 品詞タグ付け / 一致(agreement) / 統語情報 |
|---|
| Research Abstract |
The purpose of the research is to develop an analyzer capable of automatically detect grammatical errors in Spanish texts written by Japanese learners. To do that, we have manually annotated with error information a10,000 words corpus and have written a set of rules for automatic agreement error detection, using the Constraint Grammar formalism. With the evaluation of the analyzer and the annotated corpus we have confirmed that the detection of number agreement errors requires a deeper linguistic analysis of the texts than gender agreement errors, since number agreement errors occur with more frequency at the clause level than at the phrase level as the language proficiency level improves.
|
Report
(4 results)
Research Products
(16 results)
