传播结果的同源分析  

Homology analysis of diffusion results

作　　者：杨旋[1] 徐贝澄 姚畅 黄浩[2] Yang Xuan;Xu Beicheng;Yao Chang;Huang Hao(Women’s Hospital,School of Medicine,Zhejiang University,Hangzhou 310006,China;School of Computer Science,Wuhan University,Wuhan 430072,China;College of Computer Science&Technology,Zhejiang University,Hangzhou 310027,China)

机构地区：[1]浙江大学医学院附属妇产科医院,杭州310006 [2]武汉大学计算机学院,武汉430072 [3]浙江大学计算机科学与技术学院,杭州310027

出　　处：《计算机应用研究》2022年第10期2943-2949,共7页Application Research of Computers

基　　金：国家自然科学基金资助项目(61976163);浙江省博士后基金资助项目(ZJ2021017)。

摘　　要：传播结果的同源分析旨在识别出哪些历史传播结果是由同一组传播源头产生的。使用伯努利混合模型对传播结果的同源分析问题进行建模求解,同一组源头产生的传播结果对应一个伯努利分量,而伯努利分量的参数反映源头的影响在各点的传播到达概率;基于该模型和观测数据构建对数似然函数,并使用EM算法求解其中的伯努利参数,确定同源分析的结果。大量实验结果证明,对比于传统聚类算法,提出算法同源聚类的准确度更高,能够更高效准确地对传播结果数据进行同源分析。The homology analysis of diffusion results aimed to identify which historical diffusion results were produced by the same set of diffusion sources.This paper used Bernoulli mixture model to solve this problem.The diffusion results generated by the same group of sources corresponded to the same Bernoulli component,and the parameters of the corresponding Bernoulli component reflected the arrival probability of the sources’influence at each point.Then it constructed the log-likelihood function based on the model and the observed data,and used the EM algorithm to calculate the Bernoulli parameters in the log-likelihood function to determine the results of homology analysis.A large number of experimental results verify that compared with the traditional clustering algorithm,the proposed algorithm in this paper can get higher accuracy of the homologous clustering.That is to say,the proposed algorithm can perform homology analysis on the propagation result data more efficiently and accurately.

关 键 词：传播网络 同源分析 伯努利混合模型 EM算法 聚类 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]