您的位置:山东大学 -> 科技期刊社 -> 《山东大学学报(理学版)》

山东大学学报(理学版) ›› 2017, Vol. 52 ›› Issue (7): 59-65.doi: 10.6040/j.issn.1671-9352.1.2016.PC2

• • 上一篇    下一篇

基于语义图优化算法的中文微博观点摘要研究

张聪,裴家欢,黄锴宇,黄德根*,殷章志   

  1. 大连理工大学计算机科学与技术学院, 辽宁 大连 116024
  • 收稿日期:2016-11-25 出版日期:2017-07-20 发布日期:2017-07-07
  • 通讯作者: 黄德根(1965— ),男,教授,博士生导师,研究方向为自然语言处理. E-mail: huangdg@dlut.edu.cn E-mail:cccaaag@126.com
  • 作者简介:张聪(1993— ),男,硕士研究生,研究方向为自然语言处理. E-mail: cccaaag@126.com
  • 基金资助:
    国家自然科学基金资助项目(61672127)

Semantic graph optimization algorithm based chinesemicroblog opinion summarization

ZHANG Cong, PEI Jia-huan, HUANG Kai-yu, HUANG De-gen*, YIN Zhang-zhi   

  1. School of Computer Science and Technology, Dalian University of Technology, Dalian 116024, Liaoning, China
  • Received:2016-11-25 Online:2017-07-20 Published:2017-07-07

摘要: 为从海量微博中高效地获取不同话题下的关键信息,微博观点摘要成为自然语言处理领域近期研究的热点之一。基线方法基于TF-IDF算法抽取微博句中的关键词,并据此计算微博的重要性分数,直接筛选出观点摘要;朴素改进方法在基线方法的基础上,增加了情感分类步骤,并利用微博句之间的语义距离,将摘要句候选集中语义重复、重要度较小的句子去除,生成观点摘要;基于语义图优化算法的方法在朴素改进方法的基础上,利用微博句的重要性分数及微博句之间的语义距离构建语义图结构,并通过图优化算法筛选出观点摘要。朴素改进方法在COAE2016评测任务一测试数据集上,10个话题的平均ROUGE-1值达到26.39%,平均ROUGE-2值达到0.68%,平均ROUGE-SU4值达到5.69%,且评测官方公布结果显示,该方法在9项评价指标中获得6项最佳性能。基于语义图优化算法的方法在评测样例数据集上进行了实验,结果显示,该方法比朴素改进方法在ROUGE-1,ROUGE-2,ROUGE-SU4值上分别提升了0.63%, 1.51%, 2.69%。

关键词: 微博摘要, TF-IDF, 语义图优化, 句子相似度

Abstract: To obtain key information in different topics efficiently, microblog opinion summarization has been a hot spot in natural language processing recently. The baseline method of this paper extracts keywordsusing TF-IDF algorithm, and calculate the importance scores of microblogs to filter out opinion summarization directly; the naive improved methodadded a step of sentiment classification, andremove microblogs which are of low importance and high semantic repetitionusing semantic distance between microblogs to generate opinion summarization;the method based on semantic graph optimization algorithm constructs a complete graph using importance scores and semantic distance of microblogs, and filters out the opinion summarization using graph optimization algorithm. According to the official result of evaluation,on the test dataset of COAE2016, the average ROUGE-1 value, ROUGE-2 value and ROUGE-SU4 value of 10topics using the naive improved methodreached 26.39%, 0.68% and 5.69% respectively, and got 6 max values out of 9 kinds of evaluation index. Besides, the results of experiments done on COAE2016 sample datasetshows that by using the method based on semantic graph optimization algorithmthe ROUGE-1 value, ROUGE-2 value and ROUGE-SU4 value increased by 0.63%, 1.51%, 2.69% respectively.

Key words: microblogssummarization, semantic graph optimization, TF-IDF, sentence similarity

中图分类号: 

  • TP391
[1] 刘德喜, 万常选. 社会化短文本自动摘要研究综述[J]. 小型微型计算机系统, 2013, 34(12): 2764-2771. LIU Dexi, WAN Changxuan. Survey on automatic summarization of scocialized short text[J]. Journal of Chinese Computer Systems, 2013, 34(12): 2764-2771.
[2] VANDERWENDE L, SUZUKI H, BROCKETTC, et al. Beyond sumbasic: task-focused summarization with sentence simplification and lexical expansion[J]. Information Processing & Management, 2007, 43(6): 1606-1618.
[3] SINGH M, KHAN FU. Effect of incremental EM on document summarization using probabilistic latent semantic analysis[J]. Lecture Notes in Engineering and Computer Science, 2012, 2198(1):860-863.
[4] GAO D, LI W, OUYANG Y, et al. LDA-based topic formation and topic-sentence reinforcement for graph-based multi-document summarization[C] //Asia Information Retrieval Symposium. Berlin Heidelberg:Springer, 2012: 376-385.
[5] ERKAN G, RADEV D R. LexPageRank: prestige in multi-document text summarization[C].Boston: EMNLP,2004: 365-371.
[6] ZHAO L, WU L, HUANG X. Using query expansion in graph-based approach for query-focused multi-document summarization[J]. Information Processing & Management, 2009, 45(1): 35-41.
[7] WAN X, YANG J, XIAO J. Manifold-ranking based topic-focused multi-document summarization[C]. Hyderabad: IJCAI, 2007: 2903-2908.
[8] OUYANG Y, LI W, LI S, et al. Applying regression models to query-focused multi-document summarization[J]. Information Processing & Management, 2011, 47(2): 227-237.
[9] 刘晓娟, 尤斌, 张爱芸. 基于微博数据的应用研究综述[J]. 情报杂志, 2013, 32(9): 39-45. LIU Xiaojuan, YOU bin, ZHANG Aiyun, Reviewon the data used in researches of microblogs[J]. Journal of Intelligence, 2013, 32(9): 39-45.
[10] INOUYE D, KALITA J K. Comparing twitter summarization algorithms for multiple post summaries[C] // Privacy, Security, Risk and Trust(PASSAT)and 2011 IEEE Third Inernational Conference on Social Computing(SocialCom), 2011 IEEE Third International Conference on. New York:IEEE, 2011: 298-306.
[11] SHARIFI B, HUTTON M A, KALITA J. Summarizing microblogs automatically[C] // Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics. Stroudsburg: ACL, 2010: 685-688.
[12] MENG X, WEI F, LIU X, et al. Entity-centric topic-oriented opinion summarization in twitter[C] //Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM, 2012: 379-387.
[13] 林萌, 罗森林, 贾丛飞, 等. 融合句义结构模型的微博话题摘要算法[J]. 浙江大学学报(工学版), 2015(12): 2316-2325. LIN Meng, LUO Senlin, JIA Congfei, et al. MicroBlog Topics Summarization algorithm merging sentential structure model[J]. Journal of Zhejiang University(Engineering Science), 2015(12): 2316-2325.
[14] LIN C Y. Rouge: a package for automatic evaluation of summaries[C] // Text Summarization Branches Out: Proceedings of the ACL-04 Workshop. Stroudsburg: ACL, 2004, 8: 74-82.
[1] 张新猛, 蒋盛益, 张倩生, 谢柏林, 李霞. 基于用户偏好加权的混合网络推荐算法[J]. 山东大学学报(理学版), 2015, 50(09): 29-35.
[2] 王少鹏, 彭岩, 王洁. 基于LDA的文本聚类在网络舆情分析中的应用研究[J]. 山东大学学报(理学版), 2014, 49(09): 129-134.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!