• 统计研究中心
当前位置: 首页 > 系列讲座 > 正文

伊利诺伊大学香槟分校陈玉国教授:Popularity-Adjusted Block Models for Networks with Community Structure

光华讲坛——社会名流与企业家论坛第 5759 期

(线上讲座)

主题Popularity-Adjusted Block Models for Networks with Community Structure

主讲人伊利诺伊大学香槟分校陈玉国教授

主持人统计学院 林华珍教授

时间2020年7月20日(周一)上午11:00-12:00

直播平台及会议IDZOOM421 568 770

主办单位:统计研究中心、数据科学与商业智能联合实验室和统计学院 科研处

主讲人简介:

 陈玉国, 伊利诺伊大学香槟分校统计系教授,Data Science Founder Professorial Scholar,统计咨询中心主任。2001年在斯坦福大学取得统计学博士学位,他是美国统计协会(ASA)Fellow. 曾担任 Journal of American Statistical Association, Journal of Computational and Graphical Statistics, Algebraic Statistics 副主编。他的研究兴趣 包括网络数据分析,蒙特卡洛方法,和动态系统。详情请见其个人主页:https://publish.illinois.edu/yuguo/

内容提要:

The community structure observed in empirical networks has been of particular interest in the statistics literature, with a strong emphasis on the study of block models. We study an important network feature called node popularity, which is closely associated with community structure. Neither the classical stochastic block model nor its degree-corrected extension can satisfactorily capture the dynamics of node popularity as observed in empirical networks. We propose a popularity-adjusted block model for flexible and realistic modeling of node popularity. We establish consistency of likelihood modularity for community detection as well as estimation of node popularities and model parameters, and demonstrate the advantages of the new modularity over the degree-corrected block model modularity in simulations. By analyzing the political blogs network, the British MP network, and the DBLP bibliographical network, we illustrate that improved empirical insights can be gained through this methodology.

在统计文献中,人们对经验网络中观察到的社区结构特感兴趣,其中特别强调了对块模型的研究。我们研究了一个重要的网络特征,称为节点流行度,它与社区结构密切相关。正如在经验网络中观察到的那样,经典的随机块模型或其度校正的扩展都不能令人满意地捕获节点流行度的动态。我们提出了一种流行度调整的块模型,用于对节点流行度进行灵活且可践行的建模。我们建立了用于社区检测以及节点流行度和模型参数估计的一致性,并在仿真中验证了新模块优于度校正块模型的模块化。通过分析政治博客网络,英国MP网络和DBLP书目网络,我们说明可以通过这种方法获得改进的经验见解。


上一条:华东师范大学周勇教授:General assumptions in semi-supervised learning

下一条:清华大学杨朋昆博士:Towards a mathematical foundation of federated learning: a statistical perspective