查詢結果分析
來源資料
相關文獻
- The Efficient Method for Expanding Non-text Symbols Using Corpus-based Language Model in Chinese
- 護理紀錄語料及辭典建置之研究與應用於語音辨識之可行性評估
- 綜論現代中文詞語的屬性--兼介城大LIVAC共時語料庫
- 從設計簡述的語意建構產品功能模型
- Scientific Journal Articles: The Development of a Special-Purpose Corpus
- A Corps-Based Analysis of Linguistic Features in Written and Spoken English
- A Study on "Between" and "Objects of Between" in Present-Day American English Written Discourse
- Humboldt's "Spirit of the Nation" and Approaches to Its Study in Contemporary Linguistics
- 從布勒語言模型看方言田野調查
- 語料庫索引功能在英語教學上之探討
頁籤選單縮合
題 名 | The Efficient Method for Expanding Non-text Symbols Using Corpus-based Language Model in Chinese=以中文語料庫語言模型為主擴展非文字符號的有效方法 |
---|---|
作 者 | 黃豐隆; | 書刊名 | 聯合學報 |
卷 期 | 16 1999.01[民88.01] |
頁 次 | 頁241-251 |
分類號 | 802 |
關鍵詞 | 語料庫; 語言模型; 非文字符號; Corpus; Language model; Non-text symbols; |
語 文 | 英文(English) |
中文摘要 | 中央研究院平衡語料庫(ASBC)包含有 253 個不同領域文字檔案,其中有許多種 「非文字符號」, 每種符號依文句內不同文意而有多種不同口語表達( Oral Expression )。 本篇論文提出一種方法,有效解決「非文字符號」的口語歧異( Ambiguity ),預測 文句中非文字符號之類別, 再擴展成正確的口語表示式。 整個方法分成訓練階段( Training Phase )、 測試階段( Testing Phase )與擴展階段( Expanding Phase ), 以中研院平衡語料庫為基礎,建立語言模型( Language Model )。 |
英文摘要 | The corpus of Academic sinica (ASBC) includes 253 text files for different categories, within these files, some Non-text Symbols are frequently used, which can be expanded into different oral expre ssions based on related semantics in sentence. This paper addresses the method for expanding the Non-text symbols. There are two phases: training phase、 testing phase and expanaing phase. According to the Corpus (ASBC), constructing the language model, identifies its oral ambiguity and predicts its categories, then we can expand the symbols into correct oral expression. |
本系統中英文摘要資訊取自各篇刊載內容。