學術產出-Theses
Article View/Open
Publication Export
-
題名 中英文語句語意推論
Textual Entailment Recognition for Chinese and English作者 黃瑋杰
Huang, Wei Jie貢獻者 劉昭麟
Liu, Chao Lin
黃瑋杰
Huang, Wei Jie關鍵詞 語句推論
近義詞判定
經驗法則
機器學習
Entailment Recognition
Near Synonym Recognition
Heuristic Functions
Machine Learning日期 2012 上傳時間 23-Jul-2013 13:20:37 (UTC+8) 摘要 語句的推論在自然語言處理相關領域的研究，如資訊檢索、資料擷取、自動摘要或智慧型教學等，已經日趨重要。自2005年Recognizing Textual Entailment (RTE)競賽開始，此議題逐漸受到重視，而Recognizing Inference in Text (RITE-1)競賽亦開始針對中文語句推論的研究議題提供評估的平台。本研究中我們建構一個根據文本分析設計各種函式計算推論關係的模型，並提出一套基於廣義知網的詞彙語意相似度計算方法，加強推論模型對句子語意的理解能力，進而提升推論效果；此外根據過去機器學習的作法，依照上述的函式抽取詞彙語意、語法結構、POS標記、詞彙覆蓋比例與詞彙依賴關係等特徵，採用多種演算法訓練分類模型判斷推論關係。實驗結果顯示我們的兩種系統在中文語句推論關係有不錯的效能，並在NTCIR-10 RITE-2競賽中獲得第二名的佳績，同時對機器學習分類模型效能的分析也指出中英文語料於判斷推論關係時不同的特性與較有效果的特徵集。此外我們透過閱讀測驗的實驗評估，瞭解推論系統於實際應用問題的效能，並指出未來我們可以推論系統為基底，發展閱讀測驗相關的智慧型教學系統，輔助學生閱讀理解的能力與教師在閱讀測驗編輯的品質。
Recognizing Inference in Text (RITE) has become a serious issue in several research areas, such as Information Retrieval (IR), Information Extraction (IE), Automatic Summarization, or Intelligent Tutoring Systems (ITS). The research topic is getting more important since the First Recognizing Textual Entailment Challenge (RTE-1) was held in 2005. For Asian languages, Recognizing Inference in Text (RITE-1) provides evaluation standards on recognizing entailment systems. In this research, we built a system based on textual analysis and construct several heuristic functions to compute entailment in text. Besides, we proposed a method to measure the similarity between two Chinese words based on E-HowNet and used it to enhance the system’s performance. Moreover, machine learning techniques, such as SVM, J48 and Linear Regression are used to train classification models. We extracted features based on heuristic functions and other syntactic features. The experimental results indicated that our systems achieved great performances and received second places in NTCIR-10 RITE-2. The analysis of machine learning approaches also showed Chinese and English shared different linguistic characteristics and effective features on recognizing textual entailments. Besides, the experimental results of reading comprehensions showed that we can develop intelligent tutoring system based on this research. The intelligent tutoring system is able to enhance students the ability of reading understandings and help on generating quality reading tests.參考文獻 [1] 知網(HowNet)，http://www.keenage.com/[2] 重編國語辭典修訂版，http://dict.revised.moe.edu.tw/[3] 劉群、李素建，“基於《知網》的辭彙語義相似度計算”，中文計算語言學期刊，7(2)，頁59-76，2002。[4] 廣義知網知識本體架構線上瀏覽系統(Extended-HowNet) ，http://ehownet.iis.sinica.edu.tw/[5] Alexander Budanitsky and Graeme Hirst, “Semantic distance in WordNet: An experimental, application-oriented evaluation of five measures”, Workshop on WordNet and Other Lexical Resources, Second Meeting of the North American Chapter of the Association for Computational Linguistics, 2001.[6] Andrew Hickl, Jeremy Bensley, John Williams, Kirk Roberts, Bryan Rink and Ying Shi, “Recognizing Textual Entailment with LCC’s GROUNDHOG System”, Proceedings of the Second PASCAL Challenges Workshop on Recognising Textual Entailment, pp. 80-85, 2006.[7] Cheng-Wei Shih, Cheng-Wei Lee, Ting-Hao Yang and Wen-Lian Hsu. “IASL RITE System at NTCIR-9”, Proceedings of NTCIR-9 Workshop Meeting, pp. 379-385, 2011.[8] Chih-Wei Hsu, Chih-Chung Chang and Chih Jen Lin, A Practical Guide to Support Vector Classification. Retrieved from website: http://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf, 2010.[9] Chinese Knowledge Information Processing Group (CKIP), E-HowNet Technical Report. Retrieved from CKIP website: http://rocling.iis.sinica.edu.tw/CKIP/paper/Technical_Reprt_E-HowNet.pdf, 2009.[10] Chuan-Jie Lin and Bo-Yu Hsiao, “The Description of the NTOU RITE System in NTCIR-9”, Proceedings of NTCIR-9 Workshop Meeting, pp. 353-356, 2011.[11] CKIP Chinese Segmenter, http://ckipsvr.iis.sinica.edu.tw/[12] Hideki Shima, Hiroshi Kanayama, Cheng-Wei Lee, Chuan-Jie Lin, Teruko Mitamura, Yusuke Miyao, Shuming Shi and Koichi Takeda, “Overview of NTCIR-9 RITE: Recognizing Inference in TExt”, Proceedings of NTCIR-9 Workshop Meeting, pp. 291-301, 2011.[13] Ido Dagon, Oren Glickman and Bernardo Magnini, “The PASCAL Recognising Textual Entailment Challenge”, Machine Learning Challenges. Lecture Notes in Computer Science, 3944, pp. 177-190, Springer, 2006.[14] Jianfeng Gao, Mu Li, Andi Wu and Chang-Ning Huang, “Chinese Word Segmentation and Named Entity Recognition: A Pragmatic Approach”, Computational Linguistics, 31(4), 2005.[15] Kishore Papineni, Salim Roukos, Todd Ward and Wei-Jing Zhu, “BLEU: a Method for Automatic Evaluation of Machine Translation”, Proceedings of the Fortieth Annual Meeting on ACL, pp. 311-318, 2002.[16] Liang Zhou, Chin-Yew Lin and Eduard Hovy, “Re-evaluating Machine Translation Results with Paraphrase Support”, Proceedings of the Conference on EMNLP, pp. 77-84, 2006.[17] LibSVM – A Library for Support Vector Machines, http://www.csie.ntu.edu.tw/~cjlin/libsvm/[18] Ling Cao, Xipeng Qiu and Xuanjing Huang, “FudanNLP at RITE 2011: a Shallow Semantic Approach to Textual Entailment“, Proceedings of NTCIR-9 Workshop Meeting, pp. 335-338, 2011.[19] LingPipe. http://alias-i.com/lingpipe/[20] Mark Hall, Eibe Frank, Geoffrey Holmes, Bernhard Pfahringer, Peter Reutemann, Ian H. Witten, “The WEKA Data Mining Software: An Update”, SIGKDD Explorations, 11(1), 2009.[21] Marta Tatu, Brandon Iles, John Slavick, Adrian Novischi and Dan Moldovan, “COGEX at the Second Recognizing Textual Entailment Challenge”. Proceedings of the Second PASCAL Challenges Workshop on Recognising Textual Entailment, pp. 104-109, 2006.[22] Min-Yuh Day, Re-Yuan Lee, Cheng-Tai Liu, Chun Tu, Chin-Sheng Tseng, Loong Tern Yap, Allen-Green C.L. Huang, Yu-Hsuan Chiu and Wei-Ze Hong, “IMTKU Textual Entailment System for Recognizing Inference in Text at NTCIR-9 RITE”, Proceedings of NTCIR-9 Workshop Meeting, pp. 339-344, 2011.[23] Rod Adams, “Textual Entailment Through Extended Lexical Overlap”, Proceedings of the Second PASCAL Challenges Workshop on Recognising Textual Entailment, pp. 128-133, 2006.[24] Shih-Hung Wu, Wan-Chi Huang, Liang-Pu Chen and Tsun Ku, “Binary-class and Multi-class Chinese Textural Entailment System Description in NTCIR-9 RITE”, Proceedings of NTCIR-9 Workshop Meeting, pp. 422-426, 2011.[25] Stanford Dependencies, http://nlp.stanford.edu/software/stanford-dependencies.shtml[26] Stanford Named Entity Recognizer, http://www-nlp.stanford.edu/software/CRF-NER.shtml[27] Stanford Parser, http://nlp.stanford.edu/software/lex-parser.shtml[28] Stanford Tokenizer, http://nlp.stanford.edu/software/tokenizer.shtml[29] Stanford Word Segmenter, http://nlp.stanford.edu/software/segmenter.shtml[30] The Stanford Natural Language Processing Group, Stanford typed dependencies manual. Retrieved from website: http://nlp.stanford.edu/software/dependencies_manual.pdf, 2012[31] WordNet, http://wordnet.princeton.edu/[32] YAGO-NAGA Javatools, http://www.mpi-inf.mpg.de/yago-naga/javatools/[33] Yaoyun Zhang, Jun Xu, Chenlong Liu, Xiaolong Wang, Ruifeng Xu, Qingcai Chen, Xuan Wang, Yongshuai Hou and Buzhou Tang, “ICRC_HITSZ at RITE: Leveraging Multiple Classifiers Voting for Textual Entailment Recognition”, Proceedings of NTCIR-9 Workshop Meeting, pp. 325-329, 2011.[34] Yotaro Watanabe, Yusuke Miyao, Junta Mizuno, Tomohide Shibata, Hiroshi Kanayama, Cheng-Wei Lee, Chuan-Jie Lin, Shuming Shi, Teruko Mitamura, Noriko Kando, Hideki Shima and Kohichi Takeda, “Overview of the Recognizing Inference in Text (RITE-2) at NTCIR-10”, Proceedings of the Tenth NTCIR Conference, 2013. 描述 碩士
國立政治大學
資訊科學學系
100753014
101資料來源 http://thesis.lib.nccu.edu.tw/record/#G0100753014 資料類型 thesis dc.contributor.advisor 劉昭麟 zh_TW dc.contributor.advisor Liu, Chao Lin en_US dc.contributor.author (Authors) 黃瑋杰 zh_TW dc.contributor.author (Authors) Huang, Wei Jie en_US dc.creator (作者) 黃瑋杰 zh_TW dc.creator (作者) Huang, Wei Jie en_US dc.date (日期) 2012 en_US dc.date.accessioned 23-Jul-2013 13:20:37 (UTC+8) - dc.date.available 23-Jul-2013 13:20:37 (UTC+8) - dc.date.issued (上傳時間) 23-Jul-2013 13:20:37 (UTC+8) - dc.identifier (Other Identifiers) G0100753014 en_US dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/58981 - dc.description (描述) 碩士 zh_TW dc.description (描述) 國立政治大學 zh_TW dc.description (描述) 資訊科學學系 zh_TW dc.description (描述) 100753014 zh_TW dc.description (描述) 101 zh_TW dc.description.abstract (摘要) 語句的推論在自然語言處理相關領域的研究，如資訊檢索、資料擷取、自動摘要或智慧型教學等，已經日趨重要。自2005年Recognizing Textual Entailment (RTE)競賽開始，此議題逐漸受到重視，而Recognizing Inference in Text (RITE-1)競賽亦開始針對中文語句推論的研究議題提供評估的平台。本研究中我們建構一個根據文本分析設計各種函式計算推論關係的模型，並提出一套基於廣義知網的詞彙語意相似度計算方法，加強推論模型對句子語意的理解能力，進而提升推論效果；此外根據過去機器學習的作法，依照上述的函式抽取詞彙語意、語法結構、POS標記、詞彙覆蓋比例與詞彙依賴關係等特徵，採用多種演算法訓練分類模型判斷推論關係。實驗結果顯示我們的兩種系統在中文語句推論關係有不錯的效能，並在NTCIR-10 RITE-2競賽中獲得第二名的佳績，同時對機器學習分類模型效能的分析也指出中英文語料於判斷推論關係時不同的特性與較有效果的特徵集。此外我們透過閱讀測驗的實驗評估，瞭解推論系統於實際應用問題的效能，並指出未來我們可以推論系統為基底，發展閱讀測驗相關的智慧型教學系統，輔助學生閱讀理解的能力與教師在閱讀測驗編輯的品質。 zh_TW dc.description.abstract (摘要) Recognizing Inference in Text (RITE) has become a serious issue in several research areas, such as Information Retrieval (IR), Information Extraction (IE), Automatic Summarization, or Intelligent Tutoring Systems (ITS). The research topic is getting more important since the First Recognizing Textual Entailment Challenge (RTE-1) was held in 2005. For Asian languages, Recognizing Inference in Text (RITE-1) provides evaluation standards on recognizing entailment systems. In this research, we built a system based on textual analysis and construct several heuristic functions to compute entailment in text. Besides, we proposed a method to measure the similarity between two Chinese words based on E-HowNet and used it to enhance the system’s performance. Moreover, machine learning techniques, such as SVM, J48 and Linear Regression are used to train classification models. We extracted features based on heuristic functions and other syntactic features. The experimental results indicated that our systems achieved great performances and received second places in NTCIR-10 RITE-2. The analysis of machine learning approaches also showed Chinese and English shared different linguistic characteristics and effective features on recognizing textual entailments. Besides, the experimental results of reading comprehensions showed that we can develop intelligent tutoring system based on this research. The intelligent tutoring system is able to enhance students the ability of reading understandings and help on generating quality reading tests. en_US dc.description.tableofcontents 第一章 緒論 11.1 背景與動機 11.2 方法概述 21.3 成果貢獻 31.4 論文架構 3第二章 文獻回顧 52.1 中英文語句語意推論 52.2 詞彙語意相似度之計算 7第三章 語料集及中英文辭典 83.1 中文語料集 83.2 英文語料集 93.3 廣義知網(E-HowNet) 113.4 WordNet 123.5 教育部國語辭典 13第四章 研究方法 144.1 系統元件 144.1.1 數字轉換模組 144.1.2 簡繁轉換 154.1.3 中文斷詞與英文分詞 164.1.4 實體名詞(Named-Entity)標記與索引分析 164.1.5 否定詞、近義詞與反義詞搜尋 174.2 經驗法則式推論模型 224.3 機器學習之推論模型 25第五章 系統效能評估 315.1 經驗法則式推論模型實驗 315.1.1 參數調校 325.1.2 推論函式分析 345.2 機器學習式推論模型實驗 455.2.1 特徵組合搜尋 455.2.2 分類模型效能評估 485.2.3 特徵集對推論關係分類之影響 545.3 NTCIR-10競賽結果 67第六章 語句推論系統之應用 696.1 閱讀理解及語料集 696.2 文本過濾 726.3 效能分析 73第七章 結論與未來工作 787.1 結論 787.2 未來工作 81參考文獻 83附錄Ⅰ 經驗法則式推論模型參數調校及實驗 88附錄II 特徵集對推論關係分類之影響實驗 97 圖目錄圖 3.1 二元分類資料集 8圖 3.2 RTE語料範例 10圖 3.3 廣義知網詞彙定義式 11圖 3.4 廣義知網分類結構 12圖 3.5 相似詞與相反詞資料庫 13圖 4.1 原始語料之數字表達 15圖 4.2 數字轉換模組結果 15圖 4.3 實體名詞標記 17圖 4.4 專有名詞錯位 17圖 4.5 否定詞範例 18圖 4.6 否定詞辭典 18圖 4.7 近義詞範例 19圖 4.8 展開式向量形式 19圖 4.9 展開式樹狀結構 20圖 4.10 否定詞例外 21圖 4.11 反義詞範例 21圖 4.12 經驗法則式推論系統架構與流程 21圖 4.13 實體名詞位置比對範例 25圖 4.14 機器學習推論系統架構 26圖 4.15 詞彙依賴關係矩陣M 27圖 4.16 經過五步的詞彙依賴關係，M∪M2∪M3∪M4∪M5 27圖 5.1 經驗法則式推論模型近義詞效能比較：RITE-1繁體中文語料 35圖 5.2 經驗法則式推論模型近義詞效能比較：RITE-2繁體中文語料 35圖 5.3 經驗法則式推論模型近義詞效能比較：RITE-1簡體中文測試語料 37圖 5.4 經驗法則式推論模型近義詞效能比較：RITE-2簡體中文測試語料 38圖 5.5 經驗法則式推論模型系統效能：MSR測試語料 41圖 5.6 經驗法則式推論模型系統效能：RTE-1測試語料 41圖 5.7 經驗法則式推論模型系統效能：RTE-2測試語料 42圖 5.8 經驗法則式推論模型系統效能：RTE-3測試語料 42圖 5.9 分類模型準確率比較：RITE-2繁體中文訓練語料 49圖 5.10 分類模型準確率比較：RITE-2簡體中文訓練語料 49圖 5.11 分類模型準確率比較：MSR訓練語料 50圖 5.12 分類模型準確率比較：RTE訓練語料 50圖 5.13 詞彙覆蓋比例於推論關係分類之影響：RITE-1繁體中文測試語料 55圖 5.14 詞彙覆蓋比例於推論關係分類之影響：RITE-2繁體中文測試語料 55圖 5.15 詞彙覆蓋比例於推論關係分類之影響：RITE-1簡體中文測試語料 56圖 5.16 詞彙覆蓋比例於推論關係分類之影響：RITE-2簡體中文測試語料 56圖 5.17 詞彙覆蓋比例於推論關係分類之影響：MSR測試語料 57圖 5.18 詞彙覆蓋比例於推論關係分類之影響：RTE-1測試語料 57圖 5.19 詞彙覆蓋比例於推論關係分類之影響：RTE-2測試語料 58圖 5.20 詞彙覆蓋比例於推論關係分類之影響：RTE-3測試語料 58圖 5.21 近義詞於推論關係分類之影響：RITE-1繁體中文測試語料 59圖 5.22 近義詞於推論關係分類之影響：RITE-2繁體中文測試語料 60圖 5.23 近義詞於推論關係分類之影響：RITE-1簡體中文測試語料 60圖 5.24 近義詞於推論關係分類之影響：RITE-2簡體中文測試語料 61圖 5.25 語法結構於推論關係分類之影響：RITE-1繁體中文測試語料 63圖 5.26 語法結構於推論關係分類之影響：RITE-2繁體中文測試語料 63圖 5.27 語法結構於推論關係分類之影響：RITE-1簡體中文測試語料 64圖 5.28 語法結構於推論關係分類之影響：RITE-2簡體中文測試語料 64圖 5.29 語法結構於推論關係分類之影響：MSR測試語料 65圖 5.30 語法結構於推論關係分類之影響：RTE-1測試語料 65圖 5.31 語法結構於推論關係分類之影響：RTE-2測試語料 66圖 5.32 語法結構於推論關係分類之影響：RTE-3測試語料 66圖 5.33 RITE-2競賽近義詞效能比較 68圖 6.1 閱讀測驗形式 70圖 6.2 推論系統形式轉換 71圖 6.3 文章人工過濾前 72圖 6.4 文章人工過濾後 73圖 6.5 國小中文閱讀測驗透過經驗法則式推論模型實驗結果 76圖 6.6 國中英文閱讀測驗透過經驗法則式推論模型實驗結果 76圖 6.7 國小中文閱讀測驗透過機器學習分類模型實驗結果 77圖 6.8 國中英文閱讀測驗透過機器學習分類模型實驗結果 77 表目錄表 3.1 中文訓練語料集統計 9表 3.2 測試語料集數量統計 9表 3.3 MSR語料集統計 9表 3.4 RTE語料集統計 11表 4.1 中文特徵集 29表 4.2 英文特徵集 30表 5.1 RITE-2繁體中文訓練語料參數設定 33表 5.2 RITE-2簡體中文訓練語料參數設定 33表 5.3 MSR訓練語料參數設定 33表 5.4 RTE訓練語料參數設定 33表 5.5 經驗法則式推論模型實驗結果(無近義詞)：RITE-1繁體中文測試語料 36表 5.6 經驗法則式推論模型實驗結果(近義詞)：RITE-1繁體中文測試語料 36表 5.7 經驗法則式推論模型實驗結果(無近義詞)：RITE-2繁體中文測試語 37表 5.8經驗法則式推論模型實驗結果(近義詞)：RITE-2繁體中文測試語料 37表 5.9 經驗法則式推論模型實驗結果(無近義詞)：RITE-1簡體中文測試語料 39表 5.10 經驗法則式推論模型實驗結果(近義詞)：RITE-1簡體中文測試語料 39表 5.11 經驗法則式推論模型實驗結果(無近義詞)：RITE-2簡體中文測試語料 39表 5.12 經驗法則式推論模型實驗結果(近義詞)：RITE-2簡體中文測試語料 40表 5.13 經驗法則式推論模型實驗結果：MSR測試語料 43表 5.14 經驗法則式推論模型實驗結果：RTE-1測試語料 43表 5.15 經驗法則式推論模型實驗結果：RTE-2測試語料 44表 5.16 經驗法則式推論模型實驗結果：RTE-3測試語料 44表 5.17 中文特徵集編號表 46表 5.18 英文特徵集編號表 46表 5.19 RITE-2繁體中文訓練語料特徵組合搜尋 47表 5.20 RITE-2簡體中文訓練語料特徵組合搜尋 47表 5.21 MSR訓練語料特徵組合搜尋 47表 5.22 RTE訓練語料特徵組合搜尋 48表 5.23 測試語料使用之分類模型 49表 5.24 繁體中文測試語料經線性回歸演算法之推論效能 52表 5.25 簡體中文測試語料經線性回歸演算法之推論效能 52表 5.26 MSR測試語料經線性回歸演算法之推論效能 53表 5.27 RTE測試語料經SVM演算法之推論效能 53表 5.28 RITE-2繁體中文訓練語料參數設定 67表 5.29 RITE-2競賽結果(無近義詞) 67表 5.30 RITE-2競賽結果(近義詞) 67表 5.31 RITE-2中文推論模型之特徵組合- SVM 68表 5.32 RITE-2競賽結果(機器學習) 68表 6.1 閱讀測驗語料集統計 72表 6.2 閱讀測驗實驗參數設定 74表 6.3 閱讀測驗實驗特徵組合 74 zh_TW dc.format.extent 8390109 bytes - dc.format.mimetype application/pdf - dc.language.iso en_US - dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0100753014 en_US dc.subject (關鍵詞) 語句推論 zh_TW dc.subject (關鍵詞) 近義詞判定 zh_TW dc.subject (關鍵詞) 經驗法則 zh_TW dc.subject (關鍵詞) 機器學習 zh_TW dc.subject (關鍵詞) Entailment Recognition en_US dc.subject (關鍵詞) Near Synonym Recognition en_US dc.subject (關鍵詞) Heuristic Functions en_US dc.subject (關鍵詞) Machine Learning en_US dc.title (題名) 中英文語句語意推論 zh_TW dc.title (題名) Textual Entailment Recognition for Chinese and English en_US dc.type (資料類型) thesis en dc.relation.reference (參考文獻) [1] 知網(HowNet)，http://www.keenage.com/[2] 重編國語辭典修訂版，http://dict.revised.moe.edu.tw/[3] 劉群、李素建，“基於《知網》的辭彙語義相似度計算”，中文計算語言學期刊，7(2)，頁59-76，2002。[4] 廣義知網知識本體架構線上瀏覽系統(Extended-HowNet) ，http://ehownet.iis.sinica.edu.tw/[5] Alexander Budanitsky and Graeme Hirst, “Semantic distance in WordNet: An experimental, application-oriented evaluation of five measures”, Workshop on WordNet and Other Lexical Resources, Second Meeting of the North American Chapter of the Association for Computational Linguistics, 2001.[6] Andrew Hickl, Jeremy Bensley, John Williams, Kirk Roberts, Bryan Rink and Ying Shi, “Recognizing Textual Entailment with LCC’s GROUNDHOG System”, Proceedings of the Second PASCAL Challenges Workshop on Recognising Textual Entailment, pp. 80-85, 2006.[7] Cheng-Wei Shih, Cheng-Wei Lee, Ting-Hao Yang and Wen-Lian Hsu. “IASL RITE System at NTCIR-9”, Proceedings of NTCIR-9 Workshop Meeting, pp. 379-385, 2011.[8] Chih-Wei Hsu, Chih-Chung Chang and Chih Jen Lin, A Practical Guide to Support Vector Classification. Retrieved from website: http://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf, 2010.[9] Chinese Knowledge Information Processing Group (CKIP), E-HowNet Technical Report. Retrieved from CKIP website: http://rocling.iis.sinica.edu.tw/CKIP/paper/Technical_Reprt_E-HowNet.pdf, 2009.[10] Chuan-Jie Lin and Bo-Yu Hsiao, “The Description of the NTOU RITE System in NTCIR-9”, Proceedings of NTCIR-9 Workshop Meeting, pp. 353-356, 2011.[11] CKIP Chinese Segmenter, http://ckipsvr.iis.sinica.edu.tw/[12] Hideki Shima, Hiroshi Kanayama, Cheng-Wei Lee, Chuan-Jie Lin, Teruko Mitamura, Yusuke Miyao, Shuming Shi and Koichi Takeda, “Overview of NTCIR-9 RITE: Recognizing Inference in TExt”, Proceedings of NTCIR-9 Workshop Meeting, pp. 291-301, 2011.[13] Ido Dagon, Oren Glickman and Bernardo Magnini, “The PASCAL Recognising Textual Entailment Challenge”, Machine Learning Challenges. Lecture Notes in Computer Science, 3944, pp. 177-190, Springer, 2006.[14] Jianfeng Gao, Mu Li, Andi Wu and Chang-Ning Huang, “Chinese Word Segmentation and Named Entity Recognition: A Pragmatic Approach”, Computational Linguistics, 31(4), 2005.[15] Kishore Papineni, Salim Roukos, Todd Ward and Wei-Jing Zhu, “BLEU: a Method for Automatic Evaluation of Machine Translation”, Proceedings of the Fortieth Annual Meeting on ACL, pp. 311-318, 2002.[16] Liang Zhou, Chin-Yew Lin and Eduard Hovy, “Re-evaluating Machine Translation Results with Paraphrase Support”, Proceedings of the Conference on EMNLP, pp. 77-84, 2006.[17] LibSVM – A Library for Support Vector Machines, http://www.csie.ntu.edu.tw/~cjlin/libsvm/[18] Ling Cao, Xipeng Qiu and Xuanjing Huang, “FudanNLP at RITE 2011: a Shallow Semantic Approach to Textual Entailment“, Proceedings of NTCIR-9 Workshop Meeting, pp. 335-338, 2011.[19] LingPipe. http://alias-i.com/lingpipe/[20] Mark Hall, Eibe Frank, Geoffrey Holmes, Bernhard Pfahringer, Peter Reutemann, Ian H. Witten, “The WEKA Data Mining Software: An Update”, SIGKDD Explorations, 11(1), 2009.[21] Marta Tatu, Brandon Iles, John Slavick, Adrian Novischi and Dan Moldovan, “COGEX at the Second Recognizing Textual Entailment Challenge”. Proceedings of the Second PASCAL Challenges Workshop on Recognising Textual Entailment, pp. 104-109, 2006.[22] Min-Yuh Day, Re-Yuan Lee, Cheng-Tai Liu, Chun Tu, Chin-Sheng Tseng, Loong Tern Yap, Allen-Green C.L. Huang, Yu-Hsuan Chiu and Wei-Ze Hong, “IMTKU Textual Entailment System for Recognizing Inference in Text at NTCIR-9 RITE”, Proceedings of NTCIR-9 Workshop Meeting, pp. 339-344, 2011.[23] Rod Adams, “Textual Entailment Through Extended Lexical Overlap”, Proceedings of the Second PASCAL Challenges Workshop on Recognising Textual Entailment, pp. 128-133, 2006.[24] Shih-Hung Wu, Wan-Chi Huang, Liang-Pu Chen and Tsun Ku, “Binary-class and Multi-class Chinese Textural Entailment System Description in NTCIR-9 RITE”, Proceedings of NTCIR-9 Workshop Meeting, pp. 422-426, 2011.[25] Stanford Dependencies, http://nlp.stanford.edu/software/stanford-dependencies.shtml[26] Stanford Named Entity Recognizer, http://www-nlp.stanford.edu/software/CRF-NER.shtml[27] Stanford Parser, http://nlp.stanford.edu/software/lex-parser.shtml[28] Stanford Tokenizer, http://nlp.stanford.edu/software/tokenizer.shtml[29] Stanford Word Segmenter, http://nlp.stanford.edu/software/segmenter.shtml[30] The Stanford Natural Language Processing Group, Stanford typed dependencies manual. Retrieved from website: http://nlp.stanford.edu/software/dependencies_manual.pdf, 2012[31] WordNet, http://wordnet.princeton.edu/[32] YAGO-NAGA Javatools, http://www.mpi-inf.mpg.de/yago-naga/javatools/[33] Yaoyun Zhang, Jun Xu, Chenlong Liu, Xiaolong Wang, Ruifeng Xu, Qingcai Chen, Xuan Wang, Yongshuai Hou and Buzhou Tang, “ICRC_HITSZ at RITE: Leveraging Multiple Classifiers Voting for Textual Entailment Recognition”, Proceedings of NTCIR-9 Workshop Meeting, pp. 325-329, 2011.[34] Yotaro Watanabe, Yusuke Miyao, Junta Mizuno, Tomohide Shibata, Hiroshi Kanayama, Cheng-Wei Lee, Chuan-Jie Lin, Shuming Shi, Teruko Mitamura, Noriko Kando, Hideki Shima and Kohichi Takeda, “Overview of the Recognizing Inference in Text (RITE-2) at NTCIR-10”, Proceedings of the Tenth NTCIR Conference, 2013. zh_TW
