基于注意力机制的评论摘要生成

doi:10.13190/j.jbupt.2017-219

北京邮电大学学报 ›› 2018, Vol. 41 ›› Issue (3): 7-13.doi: 10.13190/j.jbupt.2017-219

基于注意力机制的评论摘要生成

苏放^1,2, 王晓宇¹, 张治^1,2

1. 北京邮电大学信息与通信工程学院, 北京 100876;
2. 无锡北邮感知技术产业研究院有限公司, 江苏无锡 214135

收稿日期:2017-11-02 出版日期:2018-06-28 发布日期:2018-06-04
作者简介:苏放(1973-),男,副教授,硕士生导师,E-mail:sufang@bupt.edu.cn.
基金资助:
国家科技重大专项项目（2017ZX03001022）

Review Summarization Generation Based on Attention Mechanism

SU Fang^1,2, WANG Xiao-yu¹, ZHANG Zhi^1,2

1. School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China;
2. Institute of Sensing Technology and Business, Beijing University of Posts and Telecommunications, Jiangsu Wuxi 214135, China

Received:2017-11-02 Online:2018-06-28 Published:2018-06-04

摘要/Abstract

摘要： 为了实现评论摘要的生成式提取，对序列到序列学习的神经网络模型进行分析，提出了一种改进的注意力机制应用模型，并用于评论摘要.挖掘评论摘要特征，使在摘要中出现的文字更多集中在原文首部；针对评论摘要的样本特征，通过改进局部注意力模型，使其对评论原文的句首具有更高的注意力权重，并可端到端地生成评论摘要的每一个词.实验结果表明，该模型在对英文同类别全文长度小于200的评论摘要提取上有更高的准确率.

关键词: 评论摘要, 注意力机制, 序列到序列, 循环神经网络

Abstract: In order to implement generative review summarization, a research of the neural network model of sequence to sequence learning was conducted, besides, an improved attention mechanism for review summarization based on the model was proposed. By focusing on the feature of review summarization samples, the local attention mechanism is improved which has more attention weights on the start of the source sentence. Then every word of the summarization is generated through the end-to-end model. Experiments show that this approach achieves superior performance on English review summarization in the same category when the length of reviews is less than 200.

Key words: review summarization, attention mechanism, sequence to sequence, recurrent neural network

中图分类号:

TN929.53

苏放, 王晓宇, 张治. 基于注意力机制的评论摘要生成[J]. 北京邮电大学学报, 2018, 41(3): 7-13.

SU Fang, WANG Xiao-yu, ZHANG Zhi. Review Summarization Generation Based on Attention Mechanism[J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2018, 41(3): 7-13.

参考文献

[1] Carbonell J, Goldstein J. The use of MMR, diversity-based reranking for reordering documents and producing summaries[C]//Proceedings of the 21^st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. New York:ACM, 1998:335-336.
[2] Conroy J M, O'leary D P. Text summarization via hidden markov models[C]//Proceedings of the 24^th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. New York:ACM, 2001:406-407.
[3] Shen D, Sun J T, Li H, et al. Document summarization using conditional random fields[C]//Proceedings of the 20^th international joint conference on Artifical intelligence. San Francisco:Morgan Kaufmann Publishers, 2007:2862-2867.
[4] Barzilay R, McKeown K R, Elhadad M. Information fusion in the context of multi-document summarization[C]//Proceedings of the 37^th Annual Meeting of the Association for Computational Linguistics on Computational Linguistics. Stroudsburg:The Association for Computational Linguistics, 1999:550-557.
[5] 李蕾, 钟义信, 郭祥昊. 面向特定领域的理解型中文自动文摘系统[J]. 计算机研究与发展, 2000(4):493-497. Li Lei, Zhong Yixin, Guo Xianghao. An understanding-based Chinese automatic abstract system in special field[J]. Journal of Computer Research and Development, 2000(4):493-497.
[6] Rush A M, Chopra S, Weston J. A neural attention model for abstractive sentence summarization[C]//Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing. Lisbon:The Association for Computational Linguistics, 2015:379-389.
[7] Firat O, Cho K, Bengio Y. Multi-way, multilingual neural machine translation with a shared attention mechanism[C]//Conference of the North American Chapter of the Association for Computational Linguistics. San Diego:Human Language Technologies, 2016:866-875.
[8] Luong M T, Pham H, Manning C D. Effective approaches to attention-based neural machine translation[C]//Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing. Lisbon:ACL, 2015:1412-1421.
[9] Sutskever I, Vinyals O, Le Q V. Sequence to sequence learning with neural networks[C]//Advances in Neural Information Processing Systems. Montreal:Curran Associates, 2014:3104-3112.
[10] Bahdanau D, Cho K, Bengio Y. Neural machine translation by jointly learning to align and translate[C]//Proceedings of the 3^rd International Conference on Learning Representations. San Diego:arXiv Press, 2015:1409. 0473V7.
[11] He R, McAuley J. Ups and downs:modeling the visual evolution of fashion trends with one-class collaborative filtering[C]//International World Wide Web Conferences Steering Committee. Proceedings of the 25^th International Conference on World Wide Web. Republic and Canton of Geneva, Switzerland:International World Wide Web Conferences Steering Committee, 2016:507-517.
[12] Chin-Yew Lin. Rouge:a package for automatic evaluation of summaries[C]//Proceedings of the ACL-04 Workshop. Barcelona:Association for Computational Linguistics, 2004:74-81.
[13] Clarke J, Lapata M. Global inference for sentence compression:an integer linear programming approach[J]. Journal of Artificial Intelligence Research, 2008, 31(1):399-429.

基于注意力机制的评论摘要生成

Review Summarization Generation Based on Attention Mechanism

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 13

编辑推荐

Metrics

本文评价

[1]	包乾辉文娟石淑珍董萌萍刘雪. 融合句法依存与 BERT-Att-BiLSTM 的属性分类[J]. 北京邮电大学学报, 2023, 46(4): 123-128.
[2]	胡敏熊思黄宏程张光华王春芳. 基于RAPNet的脑肿瘤MRI图像三维分割[J]. 北京邮电大学学报, 2023, 46(2): 91-97.
[3]	刘琳岚, 宋修洋, 陈宇斌. 基于网络表示学习的机会网络链路预测[J]. 北京邮电大学学报, 2022, 45(4): 64-69,103.
[4]	宁静, 佘红艳, 赵东, 罗丹, 王磊. 一种路网级交通事故风险预测方法[J]. 北京邮电大学学报, 2022, 45(2): 72-78.
[5]	赵海英, 朱会, 侯小刚. 基于改进EMA单元的传统服饰图像语义分割[J]. 北京邮电大学学报, 2022, 45(1): 69-74.
[6]	李剑, 陈宇航. 一种多视角高精度图片的深度估计方法[J]. 北京邮电大学学报, 2021, 44(5): 101-106.
[7]	蒲悦逸, 王文涵, 朱强, 陈朋朋. 基于CNN-ResNet-LSTM模型的城市短时交通流量预测算法[J]. 北京邮电大学学报, 2020, 43(5): 9-14.
[8]	李威, 陈曙东, 欧阳小叶, 杜蓉, 王荣. 基于高速多核网络的远监督关系抽取方法[J]. 北京邮电大学学报, 2020, 43(5): 71-76.
[9]	刘逸琛, 孙华志, 马春梅, 姜丽芬, 钟长鸿. 一种基于高层特征融合的网络商品分类[J]. 北京邮电大学学报, 2020, 43(5): 98-104,117.
[10]	王雅青, 郭彩丽, 楚云霏, 周洪弘, 冯春燕. 面向会话型推荐系统的个性化分层循环模型[J]. 北京邮电大学学报, 2019, 42(6): 142-148.
[11]	王小捷, 白子薇, 李可, 袁彩霞. 机器阅读理解的研究进展[J]. 北京邮电大学学报, 2019, 42(6): 1-9.
[12]	朱皓, 谭咏梅. 基于胶囊的英文文本蕴含识别方法[J]. 北京邮电大学学报, 2019, 42(3): 21-28.
[13]	吕光瑞, 蔡国永, 林煜明. 基于多模态判别性嵌入空间的图像情感分析[J]. 北京邮电大学学报, 2019, 42(1): 61-67.