人与人联系的矩阵分析.docVIP

人与人联系矩阵怎么得来的？？ 层次化分析得到人与人之间的层次关系及人与人之间联系概率 如何层次化分析？ 这就用到了HRG模型 模型思想：每一张图（有向、无向）G，n个节点，都对应着N张层次图(D, {pr})，这些层次图跟原图都有一定的相近度，不过不相同，我们要得到最好能描述原图的，采用的方法是极大似然估计法。得到公式如下： 符号在这里我就不一一解释，如果需要了解可以问我，英文文章中会介绍。 在这里我们用马尔可夫蒙特卡洛算法求极大似然值。通过不断迭代最后达到稳定这时的即为我们需要结果。 通过编程我们得到了人与人之间的联系矩阵，不过在这里由于篇幅原因就不贴出了。 部分翻译，可以借鉴。 Our model for the hierarchical organization of a network is as follows. Let G be a graph with n vertices. A dendrogram D is a binary tree with n leaves corresponding to the vertices of G. Each of the n?1 internal nodes of D corresponds to the group of vertices that are descended from it. We associate a probability pr with each internal node r. Then, given two vertices i, j of G, the probability pij that they are connected by an edge is pij = pr where r is their lowest common ancestor in D. The combination (D, {pr}) of the dendrogram and the set of probabilities then defines a hierarchical random graph. Assuming that all hierarchical random graphs are a priori equally likely, the probability that a given model (D, {pr}) is the correct explanation of the data is, by Bayes’ theorem, proportional to the posterior probability or likelihood L with which that model generates the observed network.3 Our goal is to maximize L or, more generally, to sample the space of all models with probability proportional to L. Let Er be the number of edges in G whose endpoints have r. as their lowest common ancestor in D, and let Lr and Rr, respectively, be the numbers of leaves in the left and right subtrees rooted at r. Then the likelihood of the hierarchical random graph is with the convention that 00 = 1. If we fix the dendrogram D, it is easy to find the probabilities {pr} that maximize L(D, {pr}). For each r, they are given by We now use a Markov chain Monte Carlo method to sample dendrograms D with probability proportional to their likelihood L(D). 通过迭代我们可以得到极大似然的HRG。 通过编程我们得到了需要的额HRG,得到Pij,从而得到人与人之间的联???矩阵。在这里由于篇幅原因我就不贴出来了。 注意:上面的翻译我是有选择摘抄的文章中的。并不是我自己写的。