摘要
为了探究网络问答社区中用户知识转移的影响要素与作用机理,本文引入社会网络分析法,构建了问答社区知识转移模型。通过分析MetaFilter问答社区人际知识网络结构,结合社会网络分析中的网络密度、结构洞、中心势、平均最短路径、聚类系数指标,揭示强关系紧密网络、强关系稀疏网络、弱关系紧密网络和弱关系稀疏网络4种不同类型的网络结构对问答社区用户知识转移的影响。分析结果表明,强关系稀疏网和弱关系紧密网对问答社区知识的有效转移具有促进作用。结构洞限制度越小,网络越稀疏,用户也会有更多的机会接触到多种异质的知识源;平均最短路径越小,小世界效应越显著,网络中知识交流更畅通;聚类系数过高或者过低都有碍于人际知识网络中知识的转移,接近于整体网络聚类系数将有助于知识的转移,并通过聚类程度指标来测量;网络中心势越大,用户在网络结构中的位置越重要,用户间的联系越紧密,对知识的转移起到显著的影响。但是,社会网络分析指标并不仅由某一个指标对知识转移起决定性作用,而是多个指标共同作用的结果,知识转移的优势网络也存在个别较低指数。探求不同网络结构对问答社区用户知识转移的影响,可以根据各指标系数调节网络结构,提高知识转移效率和效果。
To explore the influencing factors and mechanisms of user knowledge transfer in Q&A community networks,this study proposes a knowledge transfer model for Q&A communities by introducing social network analysis(SNA). This study first analyzes the interpersonal knowledge network structure of the MetaFilter Q&A community while considering net density, structural holes, central potential, average shortest path, and the clustering coefficient of SNA. The study then inspects the structures of the following four types of networks and their effects on knowledge transfer in the Q&A commu‐nity: strong-and-sparse ties, weak and tight ties, strong and tight ties, and weak and sparse ties. The results of the study in‐dicate the following.(1) Both strong-and-sparse and weak-and-tight ties networks contribute to knowledge transfer in a Q&A community.(2) When the restrictiveness of structural holes is lower and the network is sparser, users can more easily gain access to heterogeneity knowledge sources, and the average shortest path is shorter. In addition, the small-world effect is more obvious, and users can more effectively exchange knowledge. The clustering coefficient has a two-part effect in which it is neither higher nor lower. When it approximates the results of the entire network, it contributes to knowledge transfer. Furthermore, a higher central potential indicates that the relations of users are very close, and thus it has a greater influence on knowledge transfer.(3) Although the SNA index has a decisive role in knowledge transfer, some indices have a joint effect. An efficient network may still have several low indices. In summary, this study explores the influence of dif‐ferent network structures on knowledge transfer to enable network structures to be adjusted based on index coefficients,thereby improving the efficiency of knowledge transfer on Q&A communities.
作者
夏立新
杨金庆
叶光辉
程秀峰
Xia Lixin;Yang Jinqing;Ye Guanghui;Cheng Xiufeng(School of Information Management, Central China Normal University, Wuhan 430079)
出处
《情报学报》
CSSCI
CSCD
北大核心
2019年第5期447-457,共11页
Journal of the China Society for Scientific and Technical Information
基金
国家自然科学基金青年科学基金项目"基于标签语义挖掘的城市画像计算与应用模型研究"(71804055)
国家社会科学基金重大项目"基于多维度聚合的网络资源知识发现研究"(13&ZD183)
中央高校基本科研业务费项目"基于标签语义挖掘的城市画像研究"(CCNU18QN040)
关键词
问答社区
人际知识网络
知识转移
社会网络分析
Q&A community
interpersonal knowledge network
knowledge transfer
social network analysis
