Learning attribute and homophily measures through random walks
Learning attribute and homophily measures through random walks
复制标题
通过随机游走学习属性和同质性度量
DOI：
10.1007/s41109-023-00558-3
复制
发表时间：
2023
影响因子：
2.2
通讯作者：
Pipiras, Vladas
中科院分区：
文献类型：
--
作者：
Antunes, Nelson;Banerjee, Sayan;Bhamidi, Shankar;Pipiras, Vladas
研究了同伦网络中节点属性泛函的随机游动统计学习问题。属性可以是离散的，也可以是连续的。各种现有的规范模型的推广，优先附件的基础上进行了研究（模型类），新的节点形成连接依赖于他们的属性值和受欢迎程度的程度。一个相关的模型类描述，这是服从理论分析，并提供了一个主机的功能的利益渐近。分析了模型类的渐近性通过可解性现象向模型类转移的情况。对于统计学习，我们考虑了几种典型的属性不可知抽样方案，例如Metropolis-Hasting随机游走，node 2 vec的版本（Grover和Leskovec，2016），它结合了经典的随机游走和非回溯倾向，并提出了新的变体，除了拓扑信息之外，还使用属性信息来探索网络。提出了学习属性分布、属性类型的度分布和同质性测度的估计量。这样的统计学习框架的性能进行了研究，在合成网络（模型类）和真实的世界系统，其依赖于网络拓扑结构，同质性的程度或缺乏，（非）平衡的属性，进行评估。
We investigate the statistical learning of nodal attribute functionals in homophily networks using random walks. Attributes can be discrete or continuous. A generalization of various existing canonical models, based on preferential attachment is studied (model class), where new nodes form connections dependent on both their attribute values and popularity as measured by degree. An associated model classis described, which is amenable to theoretical analysis and gives access to asymptotics of a host of functionals of interest. Settings where asymptotics for model classtransfer over to model classthrough the phenomenon of resolvability are analyzed. For the statistical learning, we consider several canonical attribute agnostic sampling schemes such as Metropolis-Hasting random walk, versions of node2vec (Grover and Leskovec, 2016) that incorporate both classical random walk and non-backtracking propensities and propose new variants which use attribute information in addition to topological information to explore the network. Estimators for learning the attribute distribution, degree distribution for an attribute type and homophily measures are proposed. The performance of such statistical learning framework is studied on both synthetic networks (model class) and real world systems, and its dependence on the network topology, degree of homophily or absence thereof, (un)balanced attributes, is assessed.
登录
查看更多内容
影响因子：
3.9
作者：
SHRUM, W;CHEEK, NH;HUNTER, SM
通讯作者：
HUNTER, SM
属性引导的网络采样机制
影响因子：
3.6
作者：
Suhansanu Kumar;H. Sundaram
通讯作者：
H. Sundaram
影响因子：
--
作者：
A. Flaxman;A. Frieze;Juan C. Vera
通讯作者：
Juan C. Vera
影响因子：
2.4
作者：
Nelson Antunes;S. Bhamidi;Tianjian Guo;V. Pipiras;Bang Wang
通讯作者：
Bang Wang
识别社交网络中的时尚帐户
DOI：
--
发表时间：
2017
期刊：
影响因子：
--
作者：
D. Lee;Jinda Han;Dana Chambourova;Ranjitha Kumar
通讯作者：
Ranjitha Kumar