微博传播过程中用户影响力的特征实证分析*

摘要

在社会化信息传播媒介中，每个用户都凭借其影响力对信息传播做出贡献。利用新浪微博数据构建微博传播的完整过程，分析了用户影响力的特征及相关的因素。结果表明，用户的影响力分布极不均匀，具有双幂律分布的特征；高影响力的用户在微博传播的早期成簇出现，并且其影响力与粉丝数量之间存在一定程度的正相关性，而低影响力的用户数量巨大，影响力与粉丝数量之间没有相关性；而且，认证用户和非认证用户影响力之间存在着明显的差异。%In social media, all users make contribution to information spreading through their influences. Based on data from Sina Weibo, the complete spreading process of micro-blog is constructed;the features and correlative factors of user influence are also analyzed. Re-sults show that user influence follows an extremely heterogeneous dual power-law distribution. In the process of micro-blog spreading, high-influence users always emerge as clustering in early stage and have a positive correlation with follower numbers, while the influence of massive low-influence users are uncorrelated with follower numbers. Besides, the influence of verified users is distinctively stronger than that of unverified users.