本研究呈現片語查詢的雛形系統「PrecisePhraseBook」，能從雙語語料庫中，自動擷取英文名詞與介系詞搭配的中文翻譯及例句，可輔助語言學習者學習語言，亦能改善機器翻譯或提供語言研究者撰寫文法規則之參考。本方法利用雙語語料庫擷取英文片語的中文翻譯。其方法為使用統計方法由雙語語料庫中的詞彙自動對齊，分別擷取名詞及介系詞的翻譯，再根據由中文語料庫統計而來的中文高頻搭配詞，將名詞及介系詞的翻譯做適當調整，並產生例句。系統執行時，使用者輸入一組英文名詞與介系詞的搭配，系統會呈現資料庫中此搭配的翻譯及例句。本研究的評估方式是隨機抽取三十組名詞及介系詞的搭配，人工評估本研究方法產生的翻譯。

This thesis presents a phrases searching system, PrecisePhraseBook, which provides Chinese translations and example sentences of English phrases with a noun and preposition to assist learners in learning English or Chinese. PrecisePhraseBook provides researchers a reference tool for generating grammar rules. We propose a method for extracting Chinese translations of English phrases from bilingual parallel corpora. We use statistical methods to extract translations of nouns and prepositions from bilingual parallel corpora with sentence alignment, and then adjust the translations according to the Chinese collocations extracted from a Chinese corpus. Finally, we generate example sentences for the translations. At run-time, the user enters an English phrase with a noun and a preposition, and the system retrieves translations and example sentences from the database and presents the results to the user. The evaluation is done using randomly 30 selected phrases. We used human judge in assess the translations.

中文摘要 i
Abstract ii
致謝 iii
目錄 iv
圖次 vi
表次 vii
1 簡介 1
2 相關研究 9
3 方法 11
3.1 問題定義 11
3.2 產生翻譯 12
3.2.1 改善資料的斷詞及雙語對應 12
3.2.2 篩選實詞翻譯 16
3.2.3 篩選介系詞翻譯 33
3.2.4 產生實詞和介系詞搭配後的翻譯 41
3.3 執行階段 52
4 實驗設定及評估 53
4.1 資料集與工具 53
4.2 實驗設定 54
4.2.1 名詞介系詞搭配之片語翻譯 54
4.2.2 單詞翻譯 56
4.2.3 召回率調整 57
4.3 實驗結果與討論
5 結論與未來展望 68

Peter F. Brown, John Cocke, Stephen Della Pietra, Vincent J. Della Pietra, Fred- erick Jelinek, John D. La↵erty, Robert L. Mercer, and Paul S. Roossin. A sta- tistical approach to machine translation. Comput. Linguistics, 16:79–85, 1990.

Peter F. Brown, Jennifer C. Lai, and Robert L. Mercer. Aligning sentences in parallel corpora. In ACL, 1991.

Roberta Catizone, Graham Russell, and Susan J. Warwick. Deriving translation data from bilingual texts. 1989.

Stanley F. Chen. Aligning sentences in bilingual corpora using lexical information. In ACL, 1993.

Pascale Fung. A pattern matching method for finding noun and proper noun translations from noisy parallel corpora. In ACL, 1995.

William A. Gale and Kenneth Ward Church. Identifying word correspondences in parallel texts. In HLT, 1991.

Ming Hsien Ko. Alignment of multi-word expressions in parallel corpora. 2006. I. Dan Melamed. Automatic evaluation and uniform filter cascades for inducing n-best translation lexicons. ArXiv, cmp-lg/9505044, 1995.

Robert C. Moore. Towards a simple and accurate statistical approach to learning translation relationships among words. In DDMMT@ACL, 2001.

Michel Simard, George F. Foster, and Pierre Isabelle. Using cognates to align sentences in bilingual corpora. In CASCON, 1993.

Frank Smadja. Retrieving collocations from text: Xtract. Computational Linguistics, 19:143–177, 1993.

Dekai Wu and Xuanyin Xia. Learning an english-chinese lexicon from a parallel corpus. In AMTA, 1994.