当前位置:网站首页>复杂网络建模(三)
复杂网络建模(三)
2022-07-07 04:58:00 【坝坝头伯爵】
度中心性
度中心性分为节点度中心性和网络度中心性。前者指的是节点在其与之直接相连的邻居节点当中的中心程度,而后者则是侧重节点在整个网络的中心程度,表征的是整个网络的集中程度。
节点Vi的度中心性CD(Vi)定义为
C D ( V i ) = k i / ( N − 1 ) C_D(V_i)=k_i/(N-1) CD(Vi)=ki/(N−1)
介数中心性
介数中心性分别为节点介数中心性和网络介数中心性。
节点Vi的介数中心性CB(Vi)定义为
C B ( V i ) = 2 B i / [ ( N − 2 ) ( N − 1 ) ] C_B(V_i)=2B_i/[(N-2)(N-1)] CB(Vi)=2Bi/[(N−2)(N−1)]
接近度中心性
对于连通图来说,节点Vi的接近度中心性Cc(Vi)定义为
C c ( v i ) = ( N − 1 ) / [ ∑ j = 1 , j ≠ i N d i j ] C_c(v_i)=(N-1)/[\sum_{j=1,j\neq i}^{N}d_{ij}] Cc(vi)=(N−1)/[j=1,j=i∑Ndij]
特征向量中心性
根据网络的邻接矩阵而定义
A x = λ x Ax=\lambda x Ax=λx
只有最大的特征值对应的特征向量才是中心性测度所需要的。在得到的特征向量中,第i个分量xi就是节点Vi的特征向量中心性CE(Vi)。
有向网络的静态特征
1.入度和出度
由于与有向网络某个节点相关联的弧有指向节点的,也有背向节点向外的,因此除了可以统计与某个节点相关联的弧数(也就是度),有必要分开统计两个方向的弧数,分别成为节点的入度和出度。
加权网络的静态特征
1.点权
节点vi的点权si定义为
S j = ∑ j ∈ N i w i j S_j=\sum_{j\in N_i}^{}w_{ij} Sj=j∈Ni∑wij
式中,Ni表示节点Vi的邻点集合,wij表示连接节点vi和节点vj的边的权重。
2.单位权
节点vi的单位权Ui定义为Ui=Si/ki
3.基于节点的权-度的相关性
基于节点的权-度相关性指的是对于单个节点来说,其点权与其度之间的相关性
s ˉ ( k ) = ( ∑ i : k i = k s i ) / [ N ⋅ P ( k ) ] \bar{s}(k)=(\sum_{i:k_i=k}s_i)/[N\cdot P(k)] sˉ(k)=(i:ki=k∑si)/[N⋅P(k)]
4.权重分布差异性
节点Vi的权重分布差异性Yi表示与节点Vi相连的边权分布的离散程度,定义为
Y i = ∑ j ∈ N i ( w i j / s i ) 2 Y_i=\sum_{j\in N_i}(w_{ij}/s_i)^{2} Yi=j∈Ni∑(wij/si)2
差异性与度有如下关系:如果与节点Vi关联的边权重差异不大,则Yi正比于1/ki,如果权重差别比较大,则Yi约等于1
边栏推荐
- Paddlepaddle 29 dynamically modify the network structure without model definition code (relu changes to prelu, conv2d changes to conv3d, 2D semantic segmentation model changes to 3D semantic segmentat
- C language communication travel card background system
- Thinkcmf6.0 installation tutorial
- Network learning (I) -- basic model learning
- Force buckle 145 Binary Tree Postorder Traversal
- Recursive method to construct binary tree from preorder and inorder traversal sequence
- Merging binary trees by recursion
- JSON data flattening pd json_ normalize
- Content of string
- Common validation comments
猜你喜欢
QT learning 28 toolbar in the main window
Jmeter 的使用
Main window in QT learning 27 application
Cnopendata list data of Chinese colleges and Universities
Avatary的LiveDriver试用体验
Network learning (I) -- basic model learning
Bugku CTF daily one question chessboard with only black chess
Empire CMS collection Empire template program general
【数字IC验证快速入门】10、Verilog RTL设计必会的FIFO
【数字IC验证快速入门】14、SystemVerilog学习之基本语法1(数组、队列、结构体、枚举、字符串...内含实践练习)
随机推荐
[quickstart to Digital IC Validation] 15. Basic syntax for SystemVerilog Learning 2 (operator, type conversion, loop, Task / Function... Including practical exercises)
Introduction to basic components of wechat applet
Sign up now | oar hacker marathon phase III, waiting for your challenge
2022年茶艺师(中级)考试试题及模拟考试
2022 simulated examination question bank and online simulated examination of tea master (primary) examination questions
Quickly use Jacobo code coverage statistics
2022焊工(初级)判断题及在线模拟考试
2022制冷与空调设备运行操作复训题库及答案
Linux server development, MySQL index principle and optimization
让Livelink初始Pose与动捕演员一致
Linux server development, redis protocol and asynchronous mode
Pytorch(六) —— 模型调优tricks
Ansible
The principle and implementation of buffer playback of large video files
[VHDL parallel statement execution]
Linux server development, detailed explanation of redis related commands and their principles
dash plotly
paddlepaddle 29 无模型定义代码下动态修改网络结构(relu变prelu,conv2d变conv3d,2d语义分割模型改为3d语义分割模型)
Force buckle 144 Preorder traversal of binary tree
These five fishing artifacts are too hot! Programmer: I know, delete it quickly!