众包社区用户的知识共享网络影响机理研究

The Mechanisms of Users’ Knowledge Sharing Networkin Crowdsourcing Community

众包社区中,成员之间持续高效的知识共享行为是促进社区繁荣发展的重要因素,因此探究众包社区用户知识共享行为的影响机理具有重要意义。在收集了众包平台Kaggle中的发帖和回帖信息以及相关用户的属性数据后,构建了知识共享网络。按照竞赛的时间顺序,将用户回帖分为准备阶段、提交阶段和讨论阶段,探究不同阶段网络结构特征的演化趋势和差异,并运用ERGM从网络结构特征和用户节点属性两个角度验证影响回帖关系形成的因素。研究结果表明:在网络结构特征方面,知识共享网络具有传递性和互惠性,并且随着时间变化不断增强,优先连接机制得到验证;在用户节点属性方面,活跃度高和人气高的用户更容易获得其他用户的回帖,而地域同质性没有得到验证。

Users knowledge sharing behavior is the key factor to promote prosperity and improve development of crowdsourcing communities.Therefore,it is of great significance to explore the influence mechanisms of knowledge sharing network in crowdsourcing community.This study reveals the roles of network structure and user node attributes in forming knowledge sharing links between users in crowdsourcing community through the exponential random graph modeling approach.The data originated from a crowdsourcing platform,Kaggle.com,involve user attributes and posts.The study adopts a longitudinal design to examine how the roles of mechanisms evolve over time.In terms of roles of network structures,knowledge sharing network becomes increasingly transitive and more reciprocal over time.Users in crowdsourcing community demonstrate a preference for influential users to form knowledge sharing links.With regard to user node attributes,users have tendency to form links with those who are more active and popular.However,geographical homogeneity does not exert a significant impact on knowledge sharing link formation between users.