[Purpose/Significance] To explore the user role classification methods, key factors of role formation, transformation characteristics and differences of the Q&A platforms in the context of infectious disease outbreaks. [Method/Process] A total of 702,927 data related to Covid-19 epidemic were collected from Q&A platforms. The user roles were analyzed from the dimensions of participation and value. The influencing factors of community user role formation were constructed based on the information user factor, information factor and information environment factor. The key factors affecting the formation of different roles were analyzed by combining the multi-classification model and the SHapley Additive exPlanations (SHAP) model. The FP-growth association rule algorithm was used to mine behavior patterns and topic characteristics during the transformation of different roles. [Result/Conclusion] The results show that users tend to keep their roles unchanged, and the transformation direction is mainly towards active or diving roles. The amount of information is the key factor for the formation of different roles. There are significant differences in the extent of change in user role transformation characteristics in different transformation stages and user role transformation behaviors in all transformation stages.
[1] 人民网.习近平在中央政治局常委会会议研究应对新型冠状病毒肺炎疫情工作时的讲话[EB/OL].[2022-02-15]. http://cpc.people.com.cn/n1/2020/0215/c64094-31588554.html.
[2] 方陈承,张建同.社会化问答社区中用户研究的述评与展望[J].情报杂志,2018,37(9):185-193.
[3] 张薇薇,朱杰,蒋雪.社会学习对专业虚拟社区不同类型用户知识贡献行为的影响研究[J].情报资料工作,2021,42(5):94-103.
[4] ZHOU X, WU B, JIN Q. User role identification based on social behavior and networking analysis for information dissemination[J]. Future generation computer systems, 2019, 96:639-648.
[5] 陈烨,王乐,陈天雨,等.基于社会网络分析的社会化问答平台用户画像研究[J].情报学报,2021,40(4):414-423.
[6] HAGEL J, ARMSTRONG A. Net gain:expanding markets through virtual communities[M]. Boston:Harvard Business Press,1997.
[7] 张树森,梁循,齐金山.社会网络角色识别方法综述[J].计算机学报,2017,40(3):649-673.
[8] VILLODRE J, CRIADO J I. User roles for emergency management in social media:understanding actors'behavior during the 2018 Majorca Island flash floods[J]. Government information quarterly, 2020, 37(4):101521.
[9] PREECE J, SHNEIDERMAN B. The reader-to-leader framework:motivating technology-mediated social participation[J]. AIS transactions on human-computer interaction, 2009, 1(1):13-32.
[10] FU C. Tracking user-role evolution via topic modeling in community question answering[J]. Information processing&management, 2019, 56(6):102075.
[11] BARTAL A, RAVID G. Member behavior in dynamic online communities:role affiliation frequency model[J]. IEEE transactions on knowledge and data engineering, 2019, 32(9):1773-1784.
[12] 赵欣,王倩雯,张长征.从知识搜寻者到知识贡献者——专业虚拟社区用户角色转变的机理研究[J].情报科学,2017,35(10):18-22.
[13] ZENG G, GUAN H, CHEN F. Knowledge sharing in a virtual community of a hotel association:from free riders to active knowledge sharers[J]. Journal of China tourism research, 2014, 10(1):95-119.
[14] ERYOMIN A L. Information ecology-a viewpoint[J]. International journal of environmental studies, 1998, 54(3/4):241-253.
[15] 马海群,李钟隽,张涛.数据新闻信息生态链形成、演替及运行价值分析[J].科技情报研究,2021,3(3):49-59.
[16] NARDI B A, O'DAY V, O'DAY V L. Information ecologies:using technology with heart[M]. Cambridge:MIT Press, 1999.
[17] 娄策群,赵桂芹.信息生态平衡及其在构建和谐社会中的作用[J].情报科学,2006(11):1606-1610.
[18] 张海涛,瓮毓琦,刘阔,等.生态信息:内涵、特点、运动过程与运动规律研究[J].图书情报工作,2013,57(12):46-50.
[19] 窦悦.信息生态视角下"3×3"应急情报体系构建研究[J].图书情报工作,2020,64(15):82-89.
[20] FINK S. Crisis management:planning for the Inevitable[M]. New York:American Management Association, 1986:20.
[21] BURKHOLDER B T, TOOLE M J. Evolution of complex disasters[J]. The lancet, 1995, 346(8981):1012-1015.
[22] 安璐,陈苗苗,李纲.社交媒体环境下突发事件严重性评估和预警机制研究[J].图书情报工作,2021,65(5):98-109.
[23] 刘冰,肖高飞,晁世育.重大突发公共卫生事件风险研判与决策模型构建研究[J].信息资源管理学报, 2021,11(5):17-26,37.
[24] 姜金贵,闫思琦.基于主题和情绪相互作用的微博舆情演化研究——以"红黄蓝虐童事件"为例[J].情报杂志,2018,37(12):118-123.
[25] 安璐,吴林.融合主题与情感特征的突发事件微博舆情演化分析[J].图书情报工作,2017,61(15):120-129.
[26] 陈娟,高杉,邓胜利.社会化问答用户特征识别与行为动机分析——以"知乎"为例[J].情报科学,2017,35(5):69-74,80.
[27] 韩世曦,曾粤亮.突发公共卫生事件背景下数字青年微信公众平台健康信息采纳意愿影响因素研究[J].图书馆学研究,2021(6):83-92.
[28] 李胜利,钟滢.中外技术问答社区的实证对比研究与启示——以CSDN和Stack Overflow为例[J].情报学报,2020,39(9):989-1000.
[29] 韩盟,吴红,李昌,等.高校可转移专利识别研究——基于贝叶斯理论和组合赋权法[J].图书情报工作,2021,65(5):118-125.
[30] RIDINGS C, GEFEN D, ARINZE B. Psychological barriers:lurker and poster motivation and behavior in online communities[J]. Communications of the Association for Information Systems, 2006, 18(16):329-354.
[31] SHANNON C E. A mathematical theory of communication[J]. Bell system technical journal, 1948,27:370-423.
[32] 赵丹,王晰巍,李师萌,等.新媒体环境下的网络舆情特征量及行为规律研究——基于信息生态理论[J].情报学报,2017,36(12):1224-1232.
[33] 李嘉兴,王晰巍,李师萌,等.信息生态视角下老年用户群体微信使用行为影响因素研究[J].图书情报工作,2017,61(15):25-33.DOI:10.13266/j.issn.0252-3116.2017.15.003.
[34] TIAN H, GAO C, XIAO X, et al. SKEP:sentiment knowledge enhanced pre-training for sentiment analysis[J]. ArXiv preprint arXiv:2005.05635, 2020.
[35] Github. BERTopic[EB/OL].[2021-12-04]. https://github.com/MaartenGr/BERTopic.
[36] Github. SHAP[EB/OL].[2022-03-15]. https://github.com/slundberg/shap.
[37] HAN H, WANG W Y, MAO B H. Borderline-SMOTE:a new over-sampling method in imbalanced data sets learning[C]//International conference on intelligent computing. Berlin:Springer, 2005:878-887.
[38] BORGELT C. An implementation of the FP-growth algorithm[C]//Proceedings of the 1st international workshop on open source data mining:frequent pattern mining implementations. New York:Association for Computing Machinery, 2005:1-5.
