6.4 壓縮摘要之生成
6.4.2 候選語句評估
候選語句相當多且內容重複性大，為了從這些候選語句的集合中挑選出最為 合理重要與較多樣性(Diversified)的語句，本論文將候選語句進行兩個階段的篩選。
首先，我們透過語言模型分數將候選語句做第一次排序，排序越前面的候選語句 代表越合理重要，接著我們利用最大邊際關聯法(MMR)來剔除相似程度較高的語 句。
6.5 實驗結果
圖 6.5 節錄式正確答案摘要之範例
圖 6.6 節錄式摘要之範例
圖 6.7 壓縮摘要之範例
圖 6.8 新聞文字文件之範例
本研究以文字文件(TD)作為實驗語料，並採用本論文中具最佳效果的方法(表
5.5)作為該實驗中所使用的文件自動摘要模型，透過該模型產生出 20%的節錄式 摘要語句，並將其壓縮至 10%作為輸出的壓縮摘要。圖 6.5、6.6 與 6.7 分別屬於 新聞文字文件(圖 6.8)的節錄式正確答案之摘要、節錄式之摘要及壓縮之摘要。我 們將藉由人工選取出的正確答案摘要標記藍色，透過自動摘要系統產生的節錄式 摘要標記紅色，利用壓縮的技術產生之摘要以螢光色標記，而正確答案摘要與節 錄式摘要重疊的部分以紫色標記。
首先，從新聞文字文件(圖 6.8)中觀察到，我們使用的文件自動摘要模型所產 生的摘要與人工選取的摘要幾乎一致，證明該方法具有良好的摘要成效。而從節 錄式摘要(圖 6.6)中可以發現，由於沒有重新生成新語句，因此選取出的語句其文 法較為通順，但有過多不必要的文字資訊，將容易造成讀者閱讀不易。反觀壓縮 摘要(圖 6.7)，因利用圖學方法產生新的語句，儘管文法上有些怪異，但經過壓縮 處理後使得內容更為精簡，易於讀者閱讀且提供的摘要更多樣化(Diversified)。
第 7 章 結論與未來展望
過去在自動文件摘要的研究主要仍著重於文字文件摘要，直到 1990 年後期，
由於影音多媒體技術的進步與成熟，才逐漸開始有語音文件摘要的研究。文件摘 要可分為摘錄式摘要與抽象式摘要，本論文旨在探討節錄式中文廣播新聞文件摘 要方法。我們提出兩種詞表示法：連續型詞袋模型(CBOW)、跳躍式模型(SG)及 兩種語句表示法：分散式儲存模型(PV-DM)、分散式詞袋模型(PV-DBOW)於文件 摘要的應用，透過表示法學習的技術能將詞之間、語句之間的關聯性表現出來，
用以幫助節錄式語音文件選取重要的摘要語句。而在該技術中引入兩種加速與改 善質量的方法：階層軟式最大化(HS)與負例採樣(NS)。經由一連串的實驗分析與 討論，證明所提之方法的確可以較其它基礎實驗的摘要方法得到更高的正確率。
此外，我們除了利用文字文件的詞彙特徵及關聯特徵之外，亦利用語音訊號上之 韻律特徵對摘要的選取提供更多有幫助的資訊。
在摘要壓縮的研究中，因其一直被公認為是一個相當困難的問題，相關的研 究較少，在中文上的研究更是少見。近幾年或許因摘要壓縮在各方面的應用逐漸 受到重視，研究亦有增加的趨勢，在本論文所嘗試的摘要壓縮之基礎架構下，應 該還有更多的問題值得研究。
參考文獻
[1] H. P. Luhn, “The Automatic Creation of Literature Abstracts,” IBM Journal of Research and Development, vol. 2, no. 2, pp. 159-165, 1958.
[2] M. Mitra, A. Singhal, and C. Buckley, “Automatic Text Summarization by Paragraph Extraction,” in Proceedings of the ACL/EACL Workshop on Intelligent
Scalable Text Summarization, pp. 39-46, 1997.
[3] K. S. Jones, “Automatic Summarising: Factors and Directions,” Advances in
Automatic Text Summarization, pp. 1-12, 1999.
[4] D. Das and A. F. Martins, “A Survey on Automatic Text Summarization,”
Literature Survey for the Language and Statistics II Course at Carnegie Mellon University, vol. 4, pp. 192-195, 2007.
[5] A. Nenkova and K. McKeown, A Survey of Text Summarization Techniques, Mining Text Data, Springer, pp. 43-76, 2012.
[6] G. Tur and R. D. Mori, Spoken Language Understanding: Systems for Extracting Semantic Information from Speech, John Wiley and Sons, 2011.
[7] Y. Liu and S. Xie, “Impact of Automatic Sentence Segmentation on Meeting Summarization,” in Proceedings of the International Conference on Acoustics,
Speech and Signal Processing, pp. 5009-5012, 2008.
[8] Y. Liu, F. Liu, B. Li and S. Xie, “Do Disfluencies Affect Meeting Summarization: A Pilot Study on the Impact of Disfluencies,” in Proceedings of the Machine
Learning and Multimodal Interaction, 2007.
[9] S. Xie, Y. Liu and F. Liu, “Using N-Best Recognition Output for Extractive
Summarization and Keyword Extraction in Meeting Speech,” in Proceedings of the International Conference on Acoustics, Speech and Signal Processing, pp.
5310-5313, 2010.
[10] S. Xie and Y. Liu, “Using Confusion Networks for Speech Summarization,” in
Proceedings of the NAACL-Human Language Technologies, pp. 46-54, 2010.
[11] A. Inoue, T. Mikami and Y. Yamashita, “Improvement of Speech Summarization Using Prosodic Information,” in Proceedings of Speech Prosody, pp. 599-602, 2004.
[12] J. Zhang, “Speech Summarization Without Lexical Features for Mandarin Broadcast News,” in Proceedings of the NAACL-Human Language Technologies,
pp. 213-216, 2007.
[13] J. Zhang, H. Y. Chan, P. Fung and L. Cao, “A Comparative Study on Speech Summarization of Broadcast News and Lecture Speech,” in Proceedings of the International Speech Communication Association, pp. 2488-2491, 2007.
[14] H. Y. Chan, J. Zhang and P. Fung, “Improving Lecture Speech Summarization Using Rhetorical Information,” in Proceedings of the Automatic Speech
Recognition and Understanding, pp. 195-200, 2007.
[15] K. Knight and D. Marcu, “Summarization Beyond Sentence Extraction: A Probabilistic Approach to Sentence Compression”, in Artificial Intelligence, pp.
91-107, 2002.
[16] T. Kikuchi, S. Furui, and C. Hori, “Two-Stage Automatic Speech Summarization by Sentence Extraction and Compaction,” in Proceedings of the IEEE and ISCA
Workshop on Spontaneous Speech Processing and Recognition, pp. 207-210, 2003.
[17] S. Furui, T. Kikuchi, Y. Shinnaka and C. Hori, “Speech-to-Text and Speech-to-Speech Summarization of Spontaneous Speech”, in Proceedings of the IEEE Transactions on Speech and Audio Processing, vol. 12, no. 4, pp. 401-408,
2004.
[18] P. Baxendale, “Machine-Made Index for Technical Literature: An Experiment,”
IBM Journal of Research and Development, vol. 2, no. 4, pp. 354-361, 1958.
[19] M. Hajime and O. Manabu, “A Comparison of Summarization Methods Based on Task-based Evaluation”, in Proceedings of the International Conference on Language Resources and Evaluation, pp. 633-639, 2000.
[20] M. Hirohata, Y. Shinnaka, K. Iwano and S. Furui, “Sentence Extraction-Based Presentation Summarization Techniques and Evaluation Metrics”, in Proceedings
of the International Conference on Acoustics, Speech and Signal Processing, pp.
1065-1068, 2005.
[21] S. R. Maskey and J. Hirschberg, “Automatic Summarization of Broadcast News Using Structural Features,” in Proceedings of the European Conference on Speech Communication and Technology, pp. 1173-1176, 2003.
[22] G. Penn and X. Zhu, “A Critical Reassessment of Evaluation Baselines for Speech Summarization,” in Proceedings of the NAACL-Human Language Technologies, pp. 470-478, 2008
[23] C. Y. Lin and E. Hovy, “Automated Text Summarization and the SUMMARIST
System,” in Proceedings of the TIPSTER Text Program, pp. 197-214, 1998.
[24] G. Salton and M. E. Lesk, “Computer Evaluation of Indexing and Text
Processing,” Journal of the ACM, vol. 15, no. 1, pp. 8-36, 1968.
[25] J. Carbonell and J. Goldstein, “The Use of MMR, Diversity-Based Reranking for
Reordering Documents and Producing Summaries,” in Proceedings of the Special Interest Group on Information Retrieval, pp. 335-336, 1998.
[26] S. Xie and Y. Liu, “Using Corpus and Knowledge-Based Similarity Measure in Maximum Marginal Relevance for Meeting Summarization,” in Proceedings of
the International Conference on Acoustics, Speech and Signal Processing, pp.
4985-4988, 2008.
[27] A. Celikyilmaz and D. H. Tur, “Concept-Based Classification for Multi-Document
Summarization,” in Proceedings of the International Conference on Acoustics, Speech and Signal Processing, pp. 5540-5543, 2011.
[28] A. Celikyilmaz and D. H. Tur, “A Hybrid Hierarchical Model for Multi-Document Summarization,” in Proceedings of the Association for Computational Linguistics, pp. 815-824, 2010.
[29] Y. L. Chang and J. T. Chien, “Latent Dirichlet Learning for Document
Summarization,” in Proceedings of the International Conference on Acoustics, Speech and Signal Processing, pp. 1689-1692, 2009.
[30] Y. Gong and X. Liu, “Generic Text Summarization Using Relevance Measure and
Latent Semantic Analysis,” in Proceedings of the Special Interest Group on Information Retrieval, pp. 19-25, 2001.
[31] J. M. Ponte, and W. B. Croft, “A Language Modeling Approach to Information Retrieval,” in Proceedings of the Special Interest Group on Information Retrieval,
pp. 275-281, 1998.
[32] Y. T. Chen, B. Chen and H. M. Wang, “A Probabilistic Generative Framework for Extractive Broadcast News Speech Summarization,” IEEE Transactions on Audio, Speech and Language Processing, vol. 17, no. 1, pp. 95-106, 2009.
[33] C. X. Zhai and J. Lafferty, “A Study of Smoothing Methods for Language Models Applied to Ad Hoc Information Retrieval,” in Proceedings of the Special Interest Group on Information Retrieval, pp. 334-342, 2011.
[34] S. Kullback and R. Leibler, “On Information and Sufficiency,” The Annals of
Mathematical Statistics, vol. 22, no. 1, pp. 79-86, 1951.
[35] S. H. Lin, Y. M. Yeh and B. Chen, “Leveraging Kullback-Leibler Divergence Measures and Information-Rich Cues for Speech Summarization,” IEEE Transactions on Audio, Speech and Language Processing, vol. 19, no. 4, pp.
871-882, 2011.
[36] V. Lavrenko and W. B. Croft, “Relevance-Based Language Models,” in
Proceedings of the Special Interest Group on Information Retrieval, pp. 120-127,
2001.
[37] B. Chen, H. C. Chang, K. Y. Chen, “Sentence Modeling for Extractive Speech Summarization,” in Proceedings of the International Conference on Multimedia and Expo, pp. 1-6, 2013.
[38] S. E. Robertson, K. S. Jones, “Relevance Weighting of Search Terms,” Joumal of the American Society for Information Science, pp. 129-146, 1976.
[39] S. E. Robertson, S. Walker, S. Jones, M. Hancock-Beaulieu and M. Gatford,
“Okapi at TREC-4,” in Proceedings of the Fourth Text Retrieval Conference, pp.
73-97, 1996.
[40] S. E. Robertson and S. Walker, “Some Simple Effective Approximations to the 2-Poisson Model for Probabilistic Weighted Retrieval,” in Proceedings of the Special Interest Group on Information Retrieval, pp. 232-241, 1994.
[41] F. Rousseau and M. Vazirgiannis, “Graph-of-Word and TW-IDF: New Approach to Ad Hoc IR,” in Proceedings of the International Conference on Conference on Information, Knowledge Management, pp. 59-68, 2013.
[42] R. Motwani, L. Page, S. Brin and T. Winograd, “The PageRank Citation Ranking:
Bringing Order to the Web,” in Proceedings of the International World Wide Web Conference, pp. 161-172, 1998.
[43] C. F. Yeh, Y. N. Chen, Y. Huang and L. S. Lee, “Spoken Lecture Summarization by Random Walk over A Graph Constructed with Automatically Extracted Key Terms,” in Proceedings of the International Speech Communication Association, 2011.
[44] Y. N. Chen, H. Y. Lee and L. S. Lee, “Improved Speech Summarization and Spoken Term Detection with Graphical Analysis of Utterance Similarities,” in Proceedings of the Asia-Pacific Signal and Information Processing Association
Annual Summit and Conference, 2011.
[45] Y. N. Chen and F. Metze, “Intra-Speaker Topic Modeling for Improved Multi-Party Meeting Summarization with Integrated Random Walk,” in Proceedings of the NAACL-Human Language Technologies, pp. 377-381, 2012.
[46] Y. N. Chen and F. Metze, “Integrating Intra-Speaker Topic Modeling and Temporal-Based Inter-Speaker Topic Modeling in Random Walk for Improved Multi-Party Meeting Summarization,” in Proceedings of the International Speech Communication Association, pp. 2346-2349, 2012.
[47] Y. N. Chen and F. Metze, “Two-Layer Mutually Reinforced Random Walk for Improved Multi-Party Meeting Summarization,” in Proceedings of the Workshop on Spoken Language Technology, pp. 461-466, 2012.
[48] G. Erkan and D. R. Radev, “Lexrank: Graph-Based Centrality as Salience in Text Summarization,” Journal of Artificial Intelligence Research, vol. 22, pp. 457-479, 2004.
[49] H. Lin, J. Bilmes and S. Xie, “Graph-Based Submodular Selection for Extractive Summarization,” in Proceedings of the Automatic Speech Recognition and Understanding, pp. 381-386, 2009.
[50] N. Garg, B. Favre, K. Reidhammer and D. H. Tur, “ClusterRank: A Graph Based Method for Meeting Summarization,” in Proceedings of the International Speech Communication Association, pp. 1499-1502, 2009.
[51] C. H. Wu, C. L. Huang and C. H. Hsieh, “Spoken Document Summarization and Retrieval for Wireless Application,” in Proceedings of the Wireless Networks,
Communications, and Mobile Computing, vol. 2, pp.1388-1393, 2005.
[52] C. Y. Lin, ROUGE: Recall-Oriented Understudy for Gisting Evaluation. Available:
http://haydn.isi.edu/ROUGE/, 2003.
[53] C. Y. Lin, “ROUGE: A Package for Automatic Evaluation of Summaries,” in Proceedings of the Workshop on Text Summarization Branches Out, 2004.
[54] H. M. Wang, B. Chen, J. W. Kuo, and S. S. Cheng, “MATBN: A Mandarin Chinese Broadcast News Corpus,” Journal of Computational Linguistics and Chinese Language Processing, vol. 10, no. 2, pp. 219-236, 2005.
[55] G. E. Hinton, “Learning Distributed Representations of Concepts”, in Proceedings of the Cognitive Science Society, pp. 1-12, 1986.
[56] T. Mikolov, I. Sutskever and K. Chen, “Distributed Representations of Words and Phrases and Their Compositionality”, in Proceedings of the Neural Information Processing Systems, pp. 3111-3119, 2013.
[57] Y. Bengio, R. Ducharme, P. Vincent, and C. Jauvin, “A Neural Probabilistic Language Model,” Journal of Machine Learning Research, vol. 3, pp. 1137-1155, 2003.
[58] T. Mikolov, K. Chen, G. Corrado, and J. Dean, “Efficient Estimation of Word Representations in Vector Space,” in Proceedings of the International Conference on Learning Representations, pp. 1-12, 2013.
[59] G. Miller and W. Charles, “Contextual Correlates of Semantic Similarity,”
Language and Cognitive Processes, vol. 6, no. 1, pp. 1-28, 1991.
[60] F. Morin and Y. Bengio, “Hierarchical Probabilistic Neural Network Language Model,” in Proceedings of the Artificial Intelligence and Statistics, pp. 246-252, 2005.
[61] A. Mnih and K. Kavukcuoglu, “Learning Word Embeddings Efficiently with Noise-Contrastive Estimation,” in Proceedings of the Neural Information Processing Systems, pp. 2265-2273, 2013.
[62] J. Mitchell and M. Lapata, “Vector-Based Models of Semantic Composition,” in Proceedings of the NAACL-Human Language Technologies, pp. 236-244, 2008.
[63] F. M. Zanzotto, I. Korkontzelos, F. Fallucchi and S. Manandhar, “Estimating Linear Models for Compositional Distributional Semantics,” in Proceedings of
the International Conference on Computational Linguistics, pp. 1263-1271, 2010.
[64] A. Yessenalina and C. Cardie, “Compositional Matrix-Space Models for Sentiment Analysis,” in Proceedings of the Empirical Methods in Natural
Language Processing, pp. 172-182, 2011.
[65] Q. V. Le and T. Mikolov, “Distributed Representations of Sentences and Documents,” in Proceedings of the International Conference on Machine Learning, 2014.
[66] G. Murray, S. Renals and J. Carletta, “Extractive Summarization of Meeting Recordings,” in Proceedings of the International Speech Communication Association, pp. 593-596, 2005.
[67] J. M. Conroy and D. P. O’Leary, “Text Summarization Via Hidden Markov Models,” in Proceedings of the Special Interest Group on Information Retrieval, pp. 406-407, 2001.
[68] M. Galley, “A Skip-Chain Conditional Random Field for Ranking Meeting Utterances by Importance,” in Proceedings of the Empirical Methods in Natural Language Processing, pp. 364-372, 2006.
[69] D. Shen, J. T. Sun, H. Li, Q. Yang, and Z. Chen, “Document Summarization Using Conditional Random Fields,” in Proceedings of the International Joint Conference on Artificial Intelligence, pp. 2862-2867, 2007.
[70] J. Kupiec, “A Trainable Document Summarizer,” in Proceedings of the Special Interest Group on Information Retrieval, pp. 68-73, 1995.
[71] T. Joachims, Learning to Classify Text Using Support Vector Machines: Methods, Theory, and Algorithms, Kluwer, 2002.
[72] T. M. Cover and P. E. Hart, “Nearest Neighbor Pattern Classification,” IEEE Transactions on Information Theory, vol.13, no. 1, pp. 21-27, 1967.
[73] H. A. Rowley, S. Baluja and T. Kanade, “Neural Network-Based Face Detection,”
IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 20, no. 1,
pp. 23-38, 1998.
[74] J. R. Quinlan, “Induction of Decision Trees,” Machine Learning, vol. 1, no. 1, pp.
81-106, 1986.
[75] R. Jin and A. G. Hauptmann, “Automatic Title Generation for Spoken Broadcast News”, in Proceedings of the Human Language Conference, pp. 1-3, 2001.
[76] K. Filippova, “Multi-Sentence Compression: Finding Shortest Paths in Word Graphs ”, in Proceedings of the International Conference on Computational
Linguistics, pp. 322-330, 2010.
[77] F. Lin and Y. Liu, “From Extractive to Abstractive Meeting Summaries: Can It Be Done by Sentence Compression?,” in Proceedings of the Association for Computational Linguistics, pp. 261-264, 2009.
[78] F. Lin and Y. Liu, “Using Spoken Utterance Compression for Meeting Summarization: A Pilot Study”, in Proceedings of the Spoken Language Technology, pp. 37-42, 2010.