融合多源数据的微信用户信息分享行为特征研究

Study of WeChat Users' Information Sharing Behavior Based on Multi-source Data

【目的/意义】在于探索微信用户分享行为时空特征,为揭示用户间信息分享轨迹和社交网络结构的演化规律提供借鉴。【方法/过程】由于微信对线上数据技术采集的限制,已有的研究多集中以访谈和问卷调查的方式,对分析用户行为规律和行为特征具有很大的局限性。为了解决这一问题,本文融合多源数据采集与分析技术,实现了247711条微信用户分享数据的特征提取。【结果/结论】微信用户对微信具有较高的依赖度,分享与阅读之间的间隔时间分布具有强阵发性,而且微信中分享内容被转发的路径长度远远高于以微博为代表的综合类社交应用。

【Purpose/significance】To explore the temporal and spatial features of WeChat users’ sharing behavior, so as to reveal sharing information tracks among users and evolution regularities of social network structure.【Method/process】Due to the technology restriction on data acquisition by WeChat, most of the existing researches focused on interviews and ques-tionnaires, which had great limitations in analyzing user behavior rules and behavior characteristics. In order to solve thisproblem, this paper used the data acquisition and analysis technology to realize the feature extraction of 247711 WeChat users.【Result/conclusion】The results show that WeChat users were highly dependent on WeChat in their daily life, and theinterval time distribution between sharing users and reading users was with strongly temporal burst. Additionally, we foundthat the average path length of information shared in WeChat was much higher than that in micro-blogging.