tailieunhanh - Báo cáo khoa học: "From Extractive to Abstractive Meeting Summaries: Can It Be Done by Sentence Compression?"

Most previous studies on meeting summarization have focused on extractive summarization. In this paper, we investigate if we can apply sentence compression to extractive summaries to generate abstractive summaries. We use different compression algorithms, including integer linear programming with an additional step of filler phrase detection, a noisychannel approach using Markovization formulation of grammar rules, as well as human compressed sentences. | From Extractive to Abstractive Meeting Summaries Can It Be Done by Sentence Compression Fei Liu and Yang Liu Computer Science Department The University of Texas at Dallas Richardson TX 75080 USA feiliu yangl @ Abstract Most previous studies on meeting summarization have focused on extractive summarization. In this paper we investigate if we can apply sentence compression to extractive summaries to generate abstractive summaries. We use different compression algorithms including integer linear programming with an additional step of filler phrase detection a noisy-channel approach using Markovization formulation of grammar rules as well as human compressed sentences. Our experiments on the ICSI meeting corpus show that when compared to the abstractive summaries using sentence compression on the extractive summaries improves their ROUGE scores however the best performance is still quite low suggesting the need of language generation for abstractive summarization. 1 Introduction Meeting summaries provide an efficient way for people to browse through the lengthy recordings. Most current research on meeting summarization has focused on extractive summarization that is it extracts important sentences or dialogue acts from speech transcripts either manual transcripts or automatic speech recognition ASR output. Various approaches to extractive summarization have been evaluated recently. Popular unsupervised approaches are maximum marginal relevance MMR latent semantic analysis LSA Murray et al. 2005a and integer programming Gillick et al. 2009 . Supervised methods include hidden Markov model HMM maximum entropy conditional random fields CRF and support vector machines SVM Galley 2006 Buist et al. 2005 Xie et al. 2008 Maskey and Hirschberg 2006 . Hori et al. 2003 used a word based speech summarization approach that utilized dynamic programming to obtain a set of words to maximize a summarization score. Most of these summarization approaches aim for selecting

• Báo cáo khoa học
• From Extractive to Abstractive Meeting Summaries
• Done by Sentence Compression
• Fei Liu and Yang Liu
• báo cáo khoa học
• báo cáo ngôn ngữ
• xử lý ngôn ngữ tự nhiên
• Bài giảng Xử lý ngôn ngữ tự nhiên
• Ngôn ngữ tự nhiên
• Kỹ thuật lập trình
• Xử lý nhập nhằng ngữ nghĩa
• Xử lý ngữ nghĩa
• Xử lý ngôn ngữ
• Natural Language Processing
• Ngôn ngữ lập trình
• Nhập nhằng cấu trúc
• Tri thức về ngôn ngữ
• Mô hình ngôn ngữ
• Mô hình n gram
• Dịch máy
• Phương pháp dịch máy
• Hiểu ngôn ngữ
• Phân tích ngữ nghĩa
• Phân loại tin tự động
• Biểu diễn vị từ
• Thuộc tính về sự kiện
• Thuộc tính về sự kiện
• Mô hình ngôn ngữ Google Book N grams
• Mô hình ngôn ngữ KenLM
• Phương pháp làm mịn
• Tách từ tiếng Việt
• Gán nhãn từ loại
• Phân tích cú pháp
• Nghĩa từ vựng
• Phân giải nhập nhằng từ
• Phân loại văn bản
• Lexical chain
• Phản hồi thông tin
• Bài toán PTCP
• Cấu trúc ngữ pháp
• Phân tích cú pháp xác suất
• CKY kết hợp xác suất
• Văn phạm phi ngữ cảnh xác suất
• Hình thái học