當(dāng)今的社會(huì)網(wǎng)絡(luò),己不再是狹義上社會(huì)學(xué)研究的內(nèi)容,轉(zhuǎn)而成為了集尖端的科研價(jià)值與巨大的商業(yè)潛質(zhì)于一體的火熱研究課題,吸引著愈來愈多各領(lǐng)域的研究人員的關(guān)注。隨著時(shí)代的發(fā)展,互聯(lián)網(wǎng)中的數(shù)據(jù)也以井噴式的速度急速增加,大數(shù)據(jù)時(shí)代中的網(wǎng)絡(luò)已經(jīng)變得異常復(fù)雜。隨著逐步深入研究復(fù)雜網(wǎng)絡(luò)的物理性質(zhì)和數(shù)學(xué)特性,研究者發(fā)現(xiàn)許多真實(shí)世界的網(wǎng)絡(luò)除了具備小世界性、無標(biāo)度性這些特性外,還具有一個(gè)共同的特性,那就是社區(qū)結(jié)構(gòu),其由一系列點(diǎn)和邊組成,具有社區(qū)內(nèi)部的節(jié)點(diǎn)連接十分緊密,社區(qū)相互之間的節(jié)點(diǎn)連接松散的特征。從社區(qū)的角度能更好挖掘網(wǎng)絡(luò)的功能和價(jià)值,且便于分析網(wǎng)絡(luò)的結(jié)構(gòu)和特性。因而,挖掘出復(fù)雜網(wǎng)絡(luò)中的社區(qū)結(jié)構(gòu)具有非常重要的意義。由于缺乏將社交網(wǎng)絡(luò)轉(zhuǎn)化為數(shù)據(jù)的有效方法,有權(quán)網(wǎng)絡(luò)和無權(quán)網(wǎng)絡(luò)被當(dāng)成了兩種網(wǎng)絡(luò)分別研究,大部分對(duì)無權(quán)網(wǎng)絡(luò)的算法無法推及至權(quán)值網(wǎng)絡(luò),基于此,本文主要研究了社交網(wǎng)絡(luò)轉(zhuǎn)化為數(shù)據(jù)的方式,使得眾多應(yīng)用于數(shù)據(jù)的聚類方法可以應(yīng)用在社交網(wǎng)絡(luò)上。本文首先簡單描述了論文研究的背景、當(dāng)前的研究現(xiàn)狀和本篇論文的組織結(jié)構(gòu)。其次闡述了復(fù)雜網(wǎng)絡(luò)的含義、相關(guān)特性、拓?fù)浣Y(jié)構(gòu)模型、社區(qū)的含義,并且描述了幾種典型的社區(qū)發(fā)現(xiàn)算法。以前面的理... 

【文章來源】：西安電子科技大學(xué)陜西省 211工程院校 教育部直屬院校

【文章頁數(shù)】：76 頁

【學(xué)位級(jí)別】：碩士

【文章目錄】：
ABSTRACT
摘要
List of Symbols
List of Abbreviations
Chapter 1 Introduction
    1.1 Background
    1.2 Research status at home and abroad
    1.3 Research content
    1.4 Thesis Organization
Chapter 2 Preliminaries
    2.1 Graph Theory
    2.2 Statistical properties of complex networks
        2.2.1 Degree
        2.2.2 Neighbors Node
        2.2.3 Clustering Coefficient
        2.2.4 Shortest Path Length
        2.2.5 Density
    2.3 The Topology of the Network
        2.3.1 Scale-free Network
        2.3.2 Small-world Network
        2.3.3 Community Structure
    2.4 Classic Algorithms
        2.4.1 CMP Algorithm
        2.4.2 Label Propagation Algorithm
        2.4.3 Spectral Clustering
        2.4.4 Louvain Algorithm
    2.5 Community Evaluation Indicators
        2.5.1 Modularity
        2.5.2 Normalized Mutual Information
        2.5.3 Community in a Strong Sense and in a Weak Sense
    2.6 Conclusion
Chapter 3 Community Detection on Pseudo-Adjacency Matrix
    3.1 Problem Background and Solution
    3.2 K-means Based on Pseudo-adjacency Matrix
        3.2.1 The Maximum Degree of Initialization
        3.2.2 Basic Idea
    3.3 Hierarchical Clustering Based on Pseudo-adjacency Matrix
        3.3.1 Similarity Measure Function
        3.3.2 Basic Idea
    3.4 FCM Algorithm Based on Pseudo-adjacency Matrix
        3.4.1 Basic Knowledge
        3.4.2 Principle of Algorithm
    3.5 Conclusion
Chapter 4 Experimental Results and Analysis
    4.1 Data Set
        4.1.1 Zachary Karate Club Network
        4.1.2 Dolphins Network
        4.1.3 Football League Network
        4.1.4 Lesmis Network
    4.2 K-means Experimental Results and Analysis
        4.2.1 Choice of Parameter
        4.2.2 Comparative Experiment of Unweighted Social Network
        4.2.3 The Experiment of K-means on Weighted Network
    4.3 Hierarchical Clustering Experimental Results and Analysis
        4.3.1 Choice of Parameter
        4.3.2 Comparative Experiment of Unweighted Social Network
        4.3.3 Experiment of Hierarchical Clustering on Weighted Network
    4.4 FCM Algorithm and Experimental Analysis
        4.4.1 Performance Analysis
    4.5 Conclusion
Chapter 5 Conclusion and Future Work
    5.1 Conclusion
    5.2 Future Work
References
Acknowledgements
Biography


【參考文獻(xiàn)】：
期刊論文
[1]基于共鄰矩陣的復(fù)雜網(wǎng)絡(luò)社區(qū)結(jié)構(gòu)劃分方法[J]. 郭崇慧,張娜.  系統(tǒng)工程理論與實(shí)踐. 2010(06)
[2]大型復(fù)雜網(wǎng)絡(luò)中的社區(qū)結(jié)構(gòu)發(fā)現(xiàn)算法[J]. 胡健,董躍華,楊炳儒.  計(jì)算機(jī)工程. 2008(19)



本文編號(hào)：3198926