Tagging Simple Short Phrase in Taiwanese Using their Mandarin Counterparts in a Parallel Corpus
| Autor: | 李柏宏 |
|---|---|
| Rok vydání: | 2011 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 99 In preparing Taiwanese-Mandarin parallel corpus written in Langgeh orthography (that is, with spaces between simple short phrases), previous study explores the tagging of Mandarin simple short phrases. This paper continues the study on tagging Taiwanese simple short phrases using their Mandarin counterparts. With aim as aid to the semi-automatic tagging process, we emphasize the full correctness of tagging procedures; a partially correct tagging procedure requires manual inspection of the tagging results, an effort no less than full manual tagging. After exploring several possibilities, we come up with a simple tagging procedure that attains 100% correctness in tagging our experimental corpus, although with only 55% recall rate among all Taiwanese simple short phrases. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|