论文部分内容阅读
【目的】在科学合作网络的发展及主要社区发现方法的基础上,提出发现合作网络社区信息的方法。【方法】以情报领域部分相关期刊2012年–2016年发表论文的共著网络为实验数据,基于贝叶斯对称非负矩阵分解方法,结合自动相关确定稀疏压缩原理,实现社区数量的自动获取,并在分解过程中应用对称矩阵分解原理。【结果】通过与现有方法的比较与分析,本文方法得到较好的实验结果。【局限】网络数据获取中未引入学者甄别的优化方法。【结论】本文提出的方法能有效解决合作网络社区发现需求。
【Objective】 Based on the development of scientific cooperation network and the main methods of community discovery, this paper proposes a method of discovering community information of cooperation network. 【Method】 Based on the co-author network of some related journals in the intelligence field from 2012 to 2016, based on the Bayesian symmetric non-negative matrix factorization method and the automatic correlation to determine the principle of sparse compression, the community number was automatically obtained. Apply the principle of symmetric matrix decomposition in the decomposition process. 【Result】 Through the comparison and analysis with the existing methods, the experimental results obtained by this method are good. [Limitations] Optimization methods that have not been incorporated into scholarship in network data acquisition. [Conclusion] The method proposed in this paper can effectively solve the needs of collaborative online community discovery.