大数据环境下复杂社会网络的社区发现方法研究综述  

The Analysis on Community Detection Methods of Complex Social Networks in the Era of Big Data

作　　者：赵中英[1,2] 李超[1,2] 

机构地区：[1]山东科技大学计算机科学与工程学院 [2]山东省智慧矿山信息技术重点实验室,山东青岛266590

出　　处：《软件导刊》2016年第12期164-167,共4页Software Guide

基　　金：国家自然科学基金项目(61303167);山东科技大学人才引进启动基金项目(2015RCJJ069)

摘　　要：社会化媒体大数据环境下的社区发现研究,是社会网络分析与挖掘领域的一个热门研究方向,已有众多学者提出各种研究方法,但对当前研究工作的进展分析相对较少。首先从局部、全局、节点相似度3个角度讨论社区的定义,然后针对网络的大规模、动态、异构3个特性,分别调研与梳理国内外相关文献,并从采取的主要技术、数据建模方法、可处理的网络规模、网络时序特征4个方面比较与总结其中的代表性方法,分析当前的学术思路与发展动态,最后指出该研究领域存在的挑战及未来可能的研究方向。Community detection from big social media data is a very hot research topic in social network analysis and mining. Large number of methods have been proposed to solve the above problem. However, there still little work to make a survey or comparative analysis on those methods. In this paper, the definition of community is firstly discussed at three levels. Then the existing community detection methods are investigated in terms of three characteristics of networks., large scale, dynamic evolution and heterogeneous structure. The representative methods are compared and summarized from four aspects, which are the main adopted technologies, the data modeling method, the network size, and the temporal characteristics. The academic thinking and trends are explored based on the above work. Finally, the potential challenges and research directions in the future are pointed out.

关 键 词：大数据 社区发现 复杂社会网络 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]