淡江大學機構典藏:Item 987654321/113827
English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 62797/95867 (66%)
造訪人次 : 3731329      線上人數 : 654
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
搜尋範圍 查詢小技巧:
  • 您可在西文檢索詞彙前後加上"雙引號",以獲取較精準的檢索結果
  • 若欲以作者姓名搜尋,建議至進階搜尋限定作者欄位,可獲得較完整資料
  • 進階搜尋
    請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/113827


    題名: 機械翻訳における翻訳メモリの研究 : アライン機能を中心に
    其他題名: 機器翻譯中翻譯輔助軟體之對齊功能效用評估
    A study of translation memory softwares : alignment
    作者: 盧冠宇;Lu, Kuan-Yu
    貢獻者: 淡江大學日本語文學系碩士班
    林寄雯
    關鍵詞: 機器翻譯;翻譯對組;雙語語料庫;對齊功能;Machine Translation;Translation Pairs;Bilingual Corpus;alignment;機械翻訳;翻訳ユニット;コーパス;アライン機能
    日期: 2017
    上傳時間: 2018-08-03 14:32:44 (UTC+8)
    摘要:   本研究利用翻譯輔助軟體中之對齊功能可自動將輸入的原文及譯文文本進行對齊的特性,代入六篇不同類型文體的光華雜誌之文本進行測試。測試得出的翻譯對組有多少的正確性,使其足以利用於建構全自動翻譯資料庫之中。
      本論文由六章所構成。首先於序論介紹研究的背景和目的,以及論文的構成和研究的範圍。第一章介紹機械翻譯,以及本研究所使用的翻譯輔助軟體及對齊功能。還有用以評量對齊功能的量表。第二章介紹研究對象、研究順序和所使用的方法。第三章介紹作為研究對象的6篇文本經過對齊功能評量後的結果。第四章介紹以第三章得到的結果,對對齊失敗的翻譯對組進行整理,統計出對齊失敗的形式。第五章為結論。
     本研究舉出四個結論。一、透過對齊功能得出的翻譯對組,其中平均約有三成是達到互譯水平,可以放入全自動翻譯的資料庫中使用。二、在本研究所採用的6個樣本中,各個樣本的正確性差異很大(52.37%~4.83%)。造成此現象的原因需進行更進一步的研究。初步推測是因為文本類型的特性,以及原文作者及譯者的文風造成了影響。三、對齊功能失敗的原因大致上可分為兩種,一種為譯者的譯文有翻譯上的錯誤,或是譯文雖符合正常翻譯評價標準,但並不適合使用於全自動翻譯資料庫的翻譯對組之中。四、另一種失敗原因是對齊功能判斷錯誤造成無法對齊。
    The Alignment function of the Machine assisted human translation (MAHT) is a program that is able to automatically align the target texts with the source. In the following discussion, the Alignment function will be tested with six articles of different genres from Taiwan Panorama Magazine. The correctness of the translation pairs will be a reference for building the database of the Fully automatic high-quality translation (FAHQT).

    This thesis is consisted of six chapters. In the Introduction, the reader is provided with not only the structure of the thesis but also the background, the purpose and the field of this research. The explanation of Machine translation, and the selected MAHT, the Alignment function, and the evaluation criteria of the Alignment function will be presented in Chapter One. The object of study, the procedure and the method will be presented in Chapter Two. The testing result of the six translation pairs will be discussed in Chapter Three. In terms of the unaligned translation pairs, a further analysis will be presented in Chapter Four followed by a conclusion, Chapter Five.

    The conclusions of this study are as follows: 1. The correctness of the Alignment function is approximately 30% which can be a reference for the database of the FAHQT. 2. In terms of the percentage of the correctness, there is a major difference (52.37%~4.83%) among these six translation pairs. This difference could be caused by the genres of each article as well as the writing style of both author and translator. A further research of this difference is suggested. 3. There are two reasons for the unaligned translation pairs. The first reason is the misinterpretation of the translator. The other reason is the translation is incompatible with the database of the FAHQT despite of the fact that the translation is correct. 4. The misjudgment of the Alignment function is another cause for the unaligned translation pairs.
    顯示於類別:[日本語文學系暨研究所 ] 學位論文

    文件中的檔案:

    檔案 描述 大小格式瀏覽次數
    index.html0KbHTML133檢視/開啟

    在機構典藏中所有的資料項目都受到原著作權保護.

    TAIR相關文章

    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - 回饋