| Presentation | 2001/10/10 ATR-SLT System for SENSEVAL-2 Japanese Translation Task Tadashi KUMANO, Hideki KASHIOKA, Hideki TANAKA, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We propose a translation selection system based on the vector space model. When each translation candidate of a word is given as a pair of expressions containing the word and its translation, selecting the translation of the word can be considered equivalent to selecting the expression having the most similar context among candidate expressions. The proposed method expresses the context information in "context vectors" constructed from content words co-occurring with the target word. Context vectors represent detailed information composed of lexical attributes (word forms, semantic codes, etc.) and syntactic relations (syntactic dependency, etc.) of the co-occurring words. We tested the proposed method with the SENSEVAL-2 Japanese translation task. Precision/recall was 45.8% (49.4%, after bugelimination) to the gold standard in the experiment with the evaluation set. Furthermore, we examined the correlation between translation selection accuracy and each of lexical attribute and syntactic relation. The result shows that the semantic attributes of all context words make the greatest contribution to the performance. On the other hand, context words in some dependency relations makes contribution when they have silimar surface information, such as word forms, but make less contribution because of low recall rate. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | translation selection / disambiguation / vector space model / SENSEVAL-2 Japanese translation task |
| Paper # | NLC 2001-37 |
| Date of Issue |
| Conference Information | |
| Committee | NLC |
|---|---|
| Conference Date | 2001/10/10(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Natural Language Understanding and Models of Communication (NLC) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | ATR-SLT System for SENSEVAL-2 Japanese Translation Task |
| Sub Title (in English) | |
| Keyword(1) | translation selection |
| Keyword(2) | disambiguation |
| Keyword(3) | vector space model |
| Keyword(4) | SENSEVAL-2 Japanese translation task |
| 1st Author's Name | Tadashi KUMANO |
| 1st Author's Affiliation | ATR Spoken Language Translation Research Laboratories() |
| 2nd Author's Name | Hideki KASHIOKA |
| 2nd Author's Affiliation | ATR Spoken Language Translation Research Laboratories |
| 3rd Author's Name | Hideki TANAKA |
| 3rd Author's Affiliation | ATR Spoken Language Translation Research Laboratories |
| Date | 2001/10/10 |
| Paper # | NLC 2001-37 |
| Volume (vol) | vol.101 |
| Number (no) | 351 |
| Page | pp.pp.- |
| #Pages | 7 |
| Date of Issue | |