[目的/意义] 重大突发事件网络舆情在传播过程中往往会出现不同的主题，而微博用户对不同主题的表达和关注也会直接影响网络舆情的传播速度和规模以及舆情事件的走向。针对重大突发事件的微博用户主题演化分析有助于应急管理部门更好地理解重大突发事件的发展轨迹以及公众在不同阶段的关注点，以便采取有效应对措施。[方法/过程] 以网络舆情信息特征为立足点，辅以自然语言处理技术将舆情信息客体与本体进行剥离，结合重大突发事件特征，创新性提出以舆情客体信息为参照基线的舆情本体演化强度来反映微博主题演化趋势。[结果/结论] 研究结果表明，面向网络舆情信息本体的主题分析，与网络舆情实际发展演化趋势更加贴近，对主题内容的揭示也更加全面。同时研究思路也对现有网络舆情主题分析的研究方法中单一求助于自然语言处理技术的优化与更新具有一定启示意义。
[Purpose/Significance] The communication process of network public opinions often has different themes in major emergencies. The expression and attention of Weibo users to different topics will also directly affect the propagation speed and scale of network public opinions and the trend of public opinion events. The theme evolution analysis of Weibo users for major emergencies helps emergency management departments better understand the development trajectory of major emergencies and the public’s concerns at different stages, so as to take effective countermeasures. [Method/Process] Based on the characteristics of Internet public opinion information, this paper used natural language processing technology to separate the object and ontology of public opinion information, and combining the characteristics of major emergencies, this paper creatively proposed that the evolution strength of public opinion ontology, which took the public opinion object information as the reference baseline, to reflect the evolution trend of microblog users’ themes. [Result/Conclusion] The research results show that the theme analysis oriented to the Internet public opinion information ontology is closer to the actual development and evolution trend of Internet public opinion, and the disclosure of the theme contents is more comprehensive. At the same time, the research ideas of this paper also have some enlightenment significance for the optimization and update of existing research methods of online public opinion theme analysis that rely solely on natural language processing technology.
[1] 中华人民共和国中央人民政府. 中华人民共和国突发事件应对法[EB/OL].[2022-10-17]. http://www.gov.cn/ziliao/flfg/2007-08/30/content_732593.htm.
[2] 冯礼. 基于事件框架的突发事件信息抽取[D]. 上海:上海交通大学, 2008.
[3] KLEIN B, LAISECA X, CASADO-MANSILLA D, et al. Detection and extracting of emergency knowledge from twitter streams[C]//International conference on ubiquitous computing and ambient intelligence. Berlin:Springer, 2012:462-469.
[4] D' ANDREA E, DUCANGE P, BECHINI A, et al. Monitoring the public opinion about the vaccination topic from tweets analysis[J]. Expert systems with application, 2019, 116:209-226.
[5] 王璟琦, 李锐, 吴华意. 基于空间自相关的网络舆情话题演化时空规律分析[J]. 数据分析与知识发现, 2018, 2(2):64-73.
[6] SAKAKI T, OKAZAKI M, MATSUO Y. Earthquake shakes Twitter users:real-time event detection by social sensors[C]//Proceedings of the 19th international conference on world wide web. North Carolina:ACM, 2010:851-860.
[7] SAKAKI T, OKAZAKI M, MATSUO Y. Tweet analysis for real-time event detection and earthquake reporting system development[J]. IEEE transactions on knowledge and data engineering, 2013, 25(4):919-931.
[8] 邵琦, 牟冬梅, 王萍, 等. 基于语义的突发公共卫生事件网络舆情主题发现研究[J]. 数据分析与知识发现, 2020, 4(9):68- 80.
[9] 廖海涵, 王曰芬, 关鹏. 微博舆情传播周期中不同传播者的主题挖掘与观点识别[J]. 图书情报工作, 2018, 62(19):77-85.
[10] 黄炜, 程宝生, 杨青. 基于本体的网络群体性事件主题发现研究[J]. 图书情报工作, 2012, 56(20):47-52.
[11] 范少萍, 安新颖, 单连慧, 等. 基于医学文献的主题演化类型与演化路径识别方法研究[J]. 情报理论与实践, 2019, 42(3):114-119.
[12] 安璐, 李倩. 基于热点主题识别的突发事件次生衍生事件探测[J]. 情报资料工作, 2020, 41(6):26-35.
[13] WANG H, HOVY E, DREDZE M. The hurricane sandy Twitter corpus[C]//Proceeding of AAAI workshops at the 29th AAAI conference on artificial intelligence. Austin:AAAI Publications, 2015:20-24.
[14] 贾亚敏, 安璐, 李纲. 城市突发事件网络信息传播时序变化规律研究[J]. 情报杂志, 2015, 34(4):91-96.
[15] 安璐, 杜廷尧, 余传明, 等. 突发公共卫生事件的微博主题演化模式和时序趋势[J]. 情报资料工作, 2016(5):44-52.
[16] AN L, YU C, LIN X, et al. Topical evolution patterns and temporal trends of microblogs on public health emergencies-a case study of Ebola on Twitter and Weibo[J]. Online information review, 2018(42):821-846.
[17] MARK G, BAGDOURI M, PALEN L, et al. Blogs as a collective war diary[C]//Proceedings of the ACM 2011 conference on computer supported cooperative work. Seattle, New York:ACM, 2012:37-46.
[18] DEY L, KHURDIYA A, MAHAJAN D. Topical evolution and regional affinity of tweets[C]//Proceedings of 2013 international symposium on computational and business intelligence. Los Alamitos:IEEE, 2013:297-300.
[19] 安璐, 代园园, 周亦文. 公共安全事件衍生舆情形成与演化研究——基于话题与时间序列分析[J]. 公安学研究, 2020(1):14-31.
[20] AN L, ZHOU W J, OU M H, et al. Measuring and profiling the topical influence and sentiment contagion of public event stakeholders[J]. International journal of information management, 2021, 58(6):102327.
[21] 张柳, 王晰巍, 黄博, 等. 基于LDA模型的新冠肺炎疫情微博用户主题聚类图谱及主题传播路径研究[J]. 情报学报, 2021, 40(3):234-244.
[22] 王晰巍, 张柳, 黄博, 等. 基于LDA的微博用户主题图谱构建及实证研究——以"埃航空难"为例[J]. 数据分析与知识发现, 2020, 4(10):47-57.
[23] 王平, 谢耘耕. 突发公共事件网络舆情的形成及演变机制研究[J]. 现代传播, 2013(3):63-69.
[24] 黄微, 李瑞, 孟佳林. 大数据环境下多媒体网络舆情传播要素及运行机理研究[J]. 图书情报工作, 2015, 59(21):38-44.
[25] 王德川. 基于LDA模型的中美欧农业科研项目研究主题分析[D]. 北京:中国农业科学院, 2017.
[26] 崔凯. 基于LDA的主题演化研究与实现[D]. 长沙:国防科学技术大学, 2010.
[27] LOGUE J. Disasters, the environment, and public health:improving our response[J]. American journal of public health, 1996, 86(9):1207-1210.
[28] CALHOUN C. A world of emergencies:fear, intervention, and the limits of cosmopolitan order[J]. The Canadian review of sociology and anthropology, 2004, 41(4):373-375.
[29] EILAND J E, PRITCHARD D A, STEVENS D A. Emergency preparedness-is your or ready?[J]. AORN journal, 2004, 79(6):1275-1283.
[30] 安璐, 杜廷尧, 李纲, 等. 突发公共卫生事件利益相关者在社交媒体中的关注点及演化模式[J]. 情报学报, 2018(4):394-405.
