Presentation is loading. Please wait.

Presentation is loading. Please wait.

3.3 Network-Centric Community Detection

Similar presentations


Presentation on theme: "3.3 Network-Centric Community Detection"β€” Presentation transcript:

1 3.3 Network-Centric Community Detection
Block Model Approximation The adjacency matrix can be approximated by a block structure. Each block represents one community We approximate a given adjacency matrix A as follows: 𝑆 ∈ {0,1} π‘›Γ—π‘˜ is block indicator matrix with 𝑆 𝑖𝑗 =1 if node 𝑖 belongs to 𝑗th block. Ξ£ is π‘˜ Γ—π‘˜ matrix indicating the block (group) interaction density π‘˜ is the number of blocks

2 3.3 Network-Centric Community Detection
Block Model Approximation Natural objective is to minimize the following: The optimal S corresponds to the top k eigenvectors of the adjacency matrix A with maximum eigenvalues. K-means clustering can be applied to S to recover the community partition H .

3 3.3 Network-Centric Community Detection
Spectral Clustering It is derived from the problem of graph partition. Graph partition aims to find out a partition such that the cut is minimized. Cut - the total number of edges between two disjoint sets of nodes The green cut between two sets of nodes {1,2,3,4} and {5,6,7,8,9} is 2 Community detection problem can be reduced to finding the minimum cut in a network. This minimum cut problem can be solved efficiently It often returns imbalanced communities, with one being trivial or a singleton, i.e., a community consisting of only one node. The minimum cut is 1, between {9} and {1, 2, 3, 4, 5, 6, 7, 8} Therefore, the group sizes of communities should be considered.

4 3.3 Network-Centric Community Detection
Spectral Clustering Ratio cut & Normalized cut Graph partition: πœ‹=( 𝐢 1 , 𝐢 2 ,…, 𝐢 π‘˜ ) s.t. 𝐢 1 ∩ 𝐢 2 =βˆ… and 𝑖=1 π‘˜ 𝐢 𝑖 =𝑉 𝐢 𝑖 : the complement of 𝐢 𝑖 𝑐𝑒𝑑 𝐢 𝑖 , 𝐢 𝑖 : the number of edges between 𝐢 𝑖 and 𝐢 𝑖 π‘£π‘œπ‘™ 𝐢 𝑖 = π‘£βˆˆ 𝐢 𝑖 𝑑 𝑣 The smaller ratio/normalized cut is preferable.

5 3.3 Network-Centric Community Detection
Spectral Clustering Ratio cut & Normalized cut - Examples πœ‹ 1 =( 𝐢 1 ={9}, 𝐢 2 ={1,2,3,4,5,6,7,8}) 𝑐𝑒𝑑 𝐢 1 , 𝐢 1 =1, 𝑐𝑒𝑑 𝐢 2 , 𝐢 2 =1 𝐢 1 =1, 𝐢 2 =8 π‘£π‘œπ‘™ 𝐢 1 =1, π‘£π‘œπ‘™ 𝐢 2 =27 πœ‹ 2 =( 𝐢 1 ={1,2,3,4}, 𝐢 2 ={5,6,7,8,9}) 𝑐𝑒𝑑 𝐢 1 , 𝐢 1 =2, 𝑐𝑒𝑑 𝐢 2 , 𝐢 2 =2 𝐢 1 =4, 𝐢 2 =5 π‘£π‘œπ‘™ 𝐢 1 =12, π‘£π‘œπ‘™ 𝐢 2 =16

6 3.3 Network-Centric Community Detection
Spectral Clustering Finding the minimum ratio cut or normalized cut is NP-hard Both ratio cut and normalized cut can be formulated as a min-trace problem like below Graph Laplacian 𝐿 is defined as follows: 𝐷=π‘‘π‘–π‘Žπ‘”( 𝑑 1 , 𝑑 2 , …, 𝑑 𝑛 ) Then, 𝑆 corresponds to the top eigenvectors of 𝐿 with the smallest eigenvalues.

7 3.3 Network-Centric Community Detection
Modularity Maximization Modularity It measure the strength of a community partition for real-world networks by taking into account the degree distribution of nodes Given a network of 𝑛 nodes and π‘š edges, the expected number of edges between nodes 𝑣 𝑖 and 𝑣 𝑗 is 𝑑 𝑖 𝑑 𝑗 /2π‘š 9 nodes and 14 edges The expected number of edges between nodes 1 and 2 is 3Γ—2/(2Γ—14) = 3/14. So, 𝐴 𝑖𝑗 βˆ’ 𝑑 𝑖 𝑑 𝑗 /2π‘š measures how far the true network interaction between nodes 𝑖 and 𝑗 deviates from the expected random connections

8 3.3 Network-Centric Community Detection
Modularity Maximization Given a group of nodes 𝐢, the strength of community effect is defined as If a network is partitioned into k groups, the overall community effect can be summed up as follows Therefore, Modularity is defined as where 1/2π‘š is introduced to normalize the value between -1 and 1. Modularity calibrates the quality of community partitions thus can be used as an objective measure to maximize it

9 3.3 Network-Centric Community Detection
Modularity Maximization Define a modularity matrix 𝐡 where π‘‘βˆˆ 𝑅 𝑛×1 is a vector of each node’s degree. Let π‘†βˆˆ {0, 1} π‘›Γ—π‘˜ be a community indicator matrix with 𝑆 𝑖𝑙 =1 if node 𝑖 belongs to community 𝐢 𝑙 , and 𝑠 𝑙 the 𝑙th column of 𝑆 Modularity can be reformulated as The optimal 𝑆 can be computed as the top k eigenvectors of the modularity matrix 𝐡 with the maximum eigenvalues.


Download ppt "3.3 Network-Centric Community Detection"

Similar presentations


Ads by Google