1 / 29

SCAN: A Structural Clustering Algorithm for Networks

SCAN: A Structural Clustering Algorithm for Networks. Xiaowei Xu, Nurcan Yuruk, Zhidan Feng, and Thomas Schweiger KDD’07. An Introduction to DBSCAN. DBSCAN is a density-based algorithm. Density = number of points within a specified radius (Eps)

xandy
Download Presentation

SCAN: A Structural Clustering Algorithm for Networks

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. SCAN: A Structural Clustering Algorithm for Networks Xiaowei Xu, Nurcan Yuruk, Zhidan Feng, and Thomas Schweiger KDD’07

  2. An Introduction to DBSCAN • DBSCAN is a density-based algorithm. • Density = number of points within a specified radius (Eps) • A point is a core point if it has more than a specified number of points (MinPts) within Eps • These are points that are at the interior of a cluster • A border point has fewer than MinPts within Eps, but is in the neighborhood of a core point • A noise point is any point that is not a core point or a border point.

  3. DBSCAN: Core, Border, and Noise Points

  4. DBSCAN Algorithm • Eliminate noise points • Perform clustering on the remaining points

  5. DBSCAN: Core, Border and Noise Points Original Points Point types: core, border and noise Eps = 10, MinPts = 4

  6. Clusters When DBSCAN Works Well Original Points • Resistant to Noise • Can handle clusters of different shapes and sizes

  7. DBSCAN: Determining EPS and MinPts • Idea is that for points in a cluster, their kth nearest neighbors are at roughly the same distance • Noise points have the kth nearest neighbor at farther distance • So, plot sorted distance of every point to its kth nearest neighbor

  8. Network Clustering Problem Networks made up of the mutual relationships of data elements usually have an underlying structure. Because relationships are complex, it is difficult to discover these structures. How can the structure be made clear? Stated another way, given simply information of who associates with whom, could one identify clusters of individuals with common interests or special relationships (families, cliques, terrorist cells).

  9. An Example of Networks How many clusters? What size should they be? What is the best partitioning? Should some points be differentiated?

  10. A Social Network Model Individuals in a tight social group, or clique, know many of the same people, regardless of the size of the group. Individuals who are hubs know many people in different groups but belong to no single group. Politicians, for example bridge multiple groups. Individuals who are outliers reside at the margins of society. Hermits, for example, know few people and belong to no group.

  11. The Neighborhood of a Vertex Define () as the immediate neighborhood of a vertex (i.e. the set of people that an individual knows ).

  12. Structure Similarity The desired features tend to be captured by a measure we call Structural Similarity Structural similarity is large for members of a clique and small for hubs and outliers.

  13. Structural Connectivity [1] -Neighborhood: Core: Direct structure reachable: Structure reachable: transitive closure of direct structure reachability Structure connected: [1] M. Ester, H. P. Kriegel, J. Sander, & X. Xu (KDD'97)

  14. Structure-Connected Clusters Structure-connected cluster C Connectivity: Maximality: Hubs: Not belong to any cluster Bridge to many clusters Outliers: Not belong to any cluster Connect to less clusters hub outlier

  15. Algorithm 2 3 5 1 4 7 6 0 11 8 12 10 9 13 = 2 = 0.7

  16. Algorithm 2 3  = 2  = 0.7 5 1 4 7 6 0 11 8 12 10 9 0.63 13

  17. Algorithm 2 3  = 2  = 0.7 5 1 4 7 6 0 0.67 11 8 0.82 12 10 0.75 9 13

  18. Algorithm 2 3 5 1 4 7 6 0 11 8 12 10 9 13  = 2  = 0.7

  19. Algorithm 2 3  = 2  = 0.7 5 1 4 7 6 0 11 8 12 10 9 0.67 13

  20. Algorithm 2 3  = 2  = 0.7 5 1 4 7 6 0 11 0.73 8 0.73 12 0.73 10 9 13

  21. Algorithm 2 3  = 2  = 0.7 5 1 4 7 6 0 11 8 12 10 9 13

  22. Algorithm 2 3  = 2  = 0.7 5 1 4 7 0.51 6 0 11 8 12 10 9 13

  23. Algorithm 2 3  = 2  = 0.7 5 1 4 7 6 0 0.68 11 8 12 10 9 13

  24. Algorithm 2 3  = 2  = 0.7 5 1 4 7 6 0 11 8 12 0.51 10 9 13

  25. Algorithm 2 3  = 2  = 0.7 5 1 4 7 6 0 11 8 12 10 9 13

  26. Algorithm 2 3  = 2  = 0.7 5 1 0.51 4 7 0.68 6 0 0.51 11 8 12 10 9 13

  27. Algorithm 2 3  = 2  = 0.7 5 1 4 7 6 0 11 8 12 10 9 13

  28. Running Time Running time = O(|E|) For sparse networks = O(|V|) [2] A. Clauset, M. E. J. Newman, & C. Moore, Phys. Rev. E70, 066111 (2004).

  29. Conclusion We propose a novel network clustering algorithm: It is fast O(|E|), for scale free networks: O(|V|) It can find clusters, as well as hubs and outliers

More Related