[Purpose/Significance] The introduction of conversation analysis theory provides a new research perspective for the study of topic evolution and refines the granularity of topic evolution analysis. At the same time, a more perfect theme evolution analysis approach is applied to public emergencies, which is conducive to improving the efficiency of public opinion guidance of regulatory departments. [Method/Process] Based on the topic identification methods and topic evolution judgment criteria in existing studies, this paper combined conversation analysis and topic analysis to introduce conversation contents and conversation organization structure into the process of topic evolution analysis, and used user-generated content in COVID-19 as data source for empirical analysis. Through the topic evolution analysis based on temporal sequence and discussion hot, the evolution laws of contents at different levels were identified from the topic intensity level. The association rule calculation idea of knowledge discovery was introduced at the topic content analysis level, to mine the reference relationship between corpus contents, and the key evolution path was determined by combining the social network analysis method. [Result/Conclusion] The results show that there are certain differences in the topic contents at different levels in the network structure and it has an important influence on the evolution trend of the topic, effective supervision of the contents at important levels will play a positive role in guiding the trend of public opinion.
[1] 张宁熙.大数据在突发公共事件网络舆情信息工作中的应用[J].现代情报, 2015, 35(6):38-42.
[2] 马超,翟姗姗,王晓.会话分析视角下非正式信息交流主题与主题簇演化分析[J].图书情报工作, 2021, 65(17):91-100.
[3] RITZHAUPT A D, STEWART M, SMITH P, et al. An investigation of distance education in North American research literature using co-word analysis[J]. International review of research in open and distance learning, 2010, 11(1):37-60.
[4] 巴志超,李纲,朱世伟.共现分析中的关键词选择与语义度量方法研究[J].情报学报, 2016, 35(2):197-207.
[5] 王红斌,王健雄,张亚飞,等.主题不平衡新闻文本数据集的主题识别方法研究[J].数据分析与知识发现, 2021, 5(3):109-120.
[6] 李真,丁晟春,王楠.网络舆情观点主题识别研究[J].数据分析与知识发现, 2017, 1(8):18-30.
[7] 邱宁佳,杨长庚,王鹏,任涛.改进卷积神经网络的文本主题识别算法研究[J].计算机工程与应用,2022,58(2):161-168.
[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] 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, 37(4):394-405.
[10] 李月.突发公共卫生事件中公共政策主题演化研究——以国家中心城市官方微信为例[J].情报杂志, 2020, 39(9):143-149.
[11] 曹树金,岳文玉.突发公共卫生事件微博舆情主题挖掘与演化分析[J].信息资源管理学报, 2020, 10(6):28-37.
[12] 赵蓉英,常茹茹,陈湛,等.基于知乎平台的突发公共卫生事件主题演化研究[J].信息资源管理学报, 2021, 11(2):52-59.
[13] 姚艾昕,马捷,林英,等.重大突发公共卫生事件谣言演化与治理策略研究[J].情报科学, 2020, 38(7):22-29.
[14] 刘雅姝,张海涛,徐海玲,等.多维特征融合的网络舆情突发事件演化话题图谱研究[J].情报学报, 2019, 38(8):798-806.
[15] 徐月梅,吕思凝,蔡连侨,等.结合卷积神经网络和Topic2Vec的新闻主题演变分析[J].数据分析与知识发现, 2018, 2(9):31-41.
[16] 朱晓霞,宋嘉欣,孟建芳.基于动态主题——情感演化模型的网络舆情信息分析[J].情报科学, 2019, 37(7):72-78.
[17] 张光陆.深度学习视角下的课堂话语互动特征:基于会话分析[J].中国教育学刊, 2021(1):79-84.
[18] 王亚峰,于国栋.医患交流中患者扩展回答的会话分析研究[J].外语教学理论与实践, 2021, 175(3):108-118.
[19] 胡文芝,廖美珍.中国心理治疗话语"解述"现象的会话分析研究[J].重庆大学学报(社会科学版), 2013, 19(4):92-100.
[20] 沈芮妃.谈话节目《圆桌派》的会话结构[J].青年记者, 2017, 566(18):72-73.
[21] 张黎.现场促销员的会话策略分析[J].语言文字应用, 2007, 63(3):87-93.
[22] 巴志超,李纲,毛进,等.微信群内部信息交流的网络结构、行为及其演化分析——基于会话分析视角[J].情报学报, 2018, 37(10):1009-1021.
[23] 李纲,李显鑫,巴志超,等.微信群会话网络结构及成员角色划分研究[J].现代情报, 2018, 38(7):3-11.
[24] 卢恒,张向先,张莉曼,等.会话分析视角下虚拟学术社区用户交互行为特征研究[J].图书情报工作, 2020, 64(13):80-89.
[25] 李月琳,张建伟,张婳.螺旋式与直线式:在线健康医疗平台用户与医生交互模式研究[J].情报学报, 2021, 40(1):88-100.
[26] 桂小庆,张俊,张晓民,等.时态主题模型方法及应用研究综述[J].计算机科学, 2017, 44(2):46-55.
[27] BLEI D M, LAFFERTY J D. Dynamic topic models[C]//Proceedings of the 23rd international conference on machine learning. New York:ACM, 2006:113-120.
[28] WANG X R, MCCALLUM A. Topics over time:a non-markov continuous-time model of topical trends[C]//Proceedings of the 12th ACM SIGKDD international conference on knowledge discovery and data mining. New York:ACM, 2006:424-433.
[29] CHENG X, YAN X, LAN Y, et al. BTM:topic modeling over short texts[J]. IEEE transactions on knowledge and data engineering, 2014, 26(12):2928-2941.
[30] ZUO Y, ZHAO J, XU K. Word network topic model:a simple but general solution for short and imbalanced texts[J]. Knowledge and information systems, 2016, 48(2):379-398.
[31] 关鹏,王曰芬,傅柱.基于LDA的主题语义演化分析方法研究——以锂离子电池领域为例[J].数据分析与知识发现, 2019, 3(7):61-72.
[32] SYED S, SPRUIT M. Full-text or abstract?examining topic coherence scores using latent dirichlet allocation[C]//2017 IEEE international conference on data science and advanced analytics (DSAA). Tokyo:IEEE, 2017:165-174.
[33] RÖDER M, BOTH A, HINNEBURG A. Exploring the space of topic coherence measures[C]//Proceedings of the eighth ACM international conference on Web search and data mining. New York:ACM, 2015:399-408.
[34] 成怡,朱伟康,徐国伟.基于余弦相似度的改进ORB匹配算法[J].天津工业大学学报, 2021, 40(1):60-66.
[35] 史忠植.知识发现[M].北京:清华大学出版社, 2002.作者贡献说明:翟姗姗:确定选题;提出论文框架,论文修改;王左戎:撰写论文;陈欢:处理数据;潘港辉:获取及预处理数据。
