[目的/意义]微信群内部存在着复杂的会话网络结构，探究微信群内部网络结构特征、角色类型以及相互关系对认识微信用户信息行为与传播模式具有重要意义。[方法/过程]以真实微信群中的对话样本作为研究对象，以成员之间的@关系及交流强度作为边权重构建行为网络，采用社会网络分析和内容分析方法对微信群中会话网络结构的统计特征、核心成员识别、信息交流行为等进行分析，并设计基于数量、黏度和位置的微信群用户影响力计算模型。[结果/结论]微信群中成员发言数近似服从分段幂律分布，其幂律指数与成员活跃度成正比关系；仅依赖发言数和发言分布时间无法反映成员在网络中的地位和影响力，其与中心性、个体网络特征等指标存在不一致；微信群内的信息交流更多是一种"有限度"和"碎片化"的交流，而群间的信息流动则表现为逐级过滤、舆情风险逐级削减的趋势。
[Purpose/significance] WeChat group has complex conversational structure, and it's of far-reaching significance for understanding the WeChat users' information behavior and communication model to explore the network structure, role types and interrelation between them. [Method/process] Based on real dialogue sample as study object, the behavior network was constructed by using the@relation and communication strength between users as edge weight. The methods of social network analysis and content analysis were introduced to analysis the statistical characteristics of network structure, core member identification and information exchange behavior of in WeChat group, and an influential power model of WeChat users was designed based on volume, viscosity and position indexes.[Result/conclusion] The quantity of chats in WeChat group is approximately assumed to be in piecewise power-law distribution, and the power law index is directly proportional to user activity. Meanwhile, the quantity and time distribution of chats, which is inconsistent with the centrality, individual network characteristics and other indicators, can not reflect the users' status and influence in behavior network. Finallly, the information communication in WeChat group is more of a kind of "limited" and "fragmented" communication, and take on a trend of gradual filtering and reduction of public opinion risk between WeChat groups.
[1] 腾讯科技. 2017微信用户&生态研究报告[N/OL].[2017-04-24].http://tech.qq.com/a/20170424/004233.htm#p=1.
[2] 蒋建国.微信群:议题、身份与控制[J].探索与争鸣,2015(11):108-112.
[3] 王晰巍,曹茹烨,杨梦晴,等.微信用户信息共享行为影响因素模型及实证研究[J].图书情报工作,2016,60(15):6-13.
[4] PENG X X, ZHAO Y X, ZHU Q H. Investigating user switching intention for mobile instant messaging application:taking WeChat as an example[J].Computer in human behavior,2016,32(64):206-216.
[5] 宋雪雁,张岩琛,王小东,等.公共档案馆微信公众平台服务质量评价研究[J].图书情报工作,2016,60(16):39-49.
[6] LIEN C H, CAO Y, ZHOU X. Service quality, satisfaction, stickiness, and usage intentions:an explor-atory evaluation in the context of WeChat services[J]. Computers in human behavior, 2017,33(68):403-410.
[7] AI C, CHEN B, HE L N, et al. The geographical characteristics of WeChat propagation network[C]//Theory, methodology, tools and applications for modeling and simulation of complex systems. Singapore:Communications in computer and information science, 2016:282-292.
[8] HE J J, WANG T. Comparison and analysis of microblog and WeChat based on information spreading models[C]//Proceeding of 6th international conference on machinery, materials, environment, biotechnology and computer (MMEBC). Amsterdam:Atlantis Press, 2016:1183-1188.
[9] 史梅,翟晓娟.高校图书馆官方微信经营策略研究——以南京大学图书馆为例[J].大学图书馆学报,2014,32(5):79-85.
[10] 方卿.论网络载体环境下科学信息交流过程的基本特征[J].情报理论与实践,2002,25(2):95-98.
[11] 王知津,李新华.论信息网络化对传统情报交流模式的影响[J].图书情报工作,1999,43(1):8-10.
[12] 黄水清,王超.网络环境下非正式信息交流的概念、类别与特点[J].图书馆杂志,2004,23(6):8-11.
[13] 马费成.Web2.0信息自组织功能与序化机制研究-兼论情报学的有序性原理[J].图书情报知识,2008(3):5-7.
[14] 姜永常.基于用户体验的知识构建——Web2.0环境下对知识构建原理的再认识[J].情报学报,2010,29(5):872-879.
[15] 刘春茂,米国伟.基于Web2.0的网络用户群"社会性"行为的系统分析[J].图书情报工作,2010,54(20):32-35.
[16] YUAN W G, LIU Y. A mixing evolution model for bidirectional microblog user networks[J]. Physica A:statistical mechanics and its applications,2015,432(8):167-179.
[17] CAPOCCI A, SERVEDIO V D, COLAIORI F, et al. Preferential attachment in the growth of social networks:the internet encyclopedia wikipedia[J]. Physical review E, 2006,74(3):036116.
[18] BARABASI A L. The origin of bursts and heavy tails in human dynamics[J].Nature,2005, 435(7039):207-211.
[19] ELLISON N, STEINFIELD C, LAMPE C. Spatially bounded online social networks and social capital[J]. International communication association, 2006,18(13):1-37.
[20] LESKOVEC J, HORVITZ E. Planetary-scale views on a large instant-messaging Network[C]//Proceeding of the 17th international conference on World Wide Web. New York:ACM Press,2008:915-924.
[21] 王芳,翟羽佳.微信群社会结构及其演化:基于文本挖掘的案例分析[J].情报学报,2016,35(6):617-629.
[22] KITSAK M, GALLONS L, HAVLIN S, et al. Identification of influential spreaders in complex networks[J]. Nature physics,2010,6(11):888-893.
[23] ABBASI A. h-Type hybrid centrality measures for weighted network[J]. Scientometrics, 2013, 96(2):633-640.
[24] YAN E. DING Y. Applying centrality measures to impact analysis:a coauthorship network analysis[J]. Journal of the American Society for Information Science and Technology,2009,60(10):2017-2118.
[25] CHEN D B, LU L Y, SHANG M S, et al. Identifying influential nodes in complex network[J]. Physica A:statistical mechanics and its applications,2012,391(4):1777-1787.
[26] 周漩,张凤鸣,李克武,等.利用重要度评价矩阵确定复杂网络关键节点[J].物理学报,2012,61(5):50201-50201
[27] HUANG X, VODENSKA I, WANG F, et al. Identifying influential directors in the united states corporate governance network[J]. Physical review E,2011,84(4):046101.
[28] SABIDUSSI G. The centrality index of a graph[J]. Psychometrika,1966,31(4):581-603.
[29] ZHAO S X. ROUSEAU R, YE F Y. h-Degree as a basic measure as a basic measure in weighted network[J]. Journal of informetrics,2011,5(4):668-677.
[30] HIRSCH J E. An index to quantify an individual's scientific research output[J]. Proceeding of the National Academy of Sciences of the United States of America, 2005,102(46):16569-16572.
