Download presentation

Presentation is loading. Please wait.

Published byDwayne Durham Modified over 2 years ago

1
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax K-MST -based clustering Caiming Zhong Pasi Franti

2
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax Outline Minimum spanning tree (MST) MST-based clustering K-MST K-MST-based clustering Fast approximate MST MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

3
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax Minimum Spanning Tree Spanning tree Given graph Spanning tree Non- Spanning tree MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

4
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax Minimum Spanning Tree Minimize the sum of weights (Kruskal, Prim’s Algorithm) Given graph G=(V,E) MST T MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

5
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax MST-based clustering The most used Method1: removing long MST-edges MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

6
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

7
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax MST-based clustering Removing long MST-edges doesn’t always work MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

8
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax MST-based clustering The most used Method2: edge inconsistent Tree edge AB, whose weight W(AB) is significantly larger than the average of nearby edge weights on both sides of the edge AB, should be deleted. MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

9
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax K-MST What is K-MST? –Let G = (V,E) denote the complete graph –Let MST 1 denote the MST of G, and it is computed as MST 1 = mst(V, E). –Then, MST 2 denote the second round of MST of G, MST 2 = mst(V, E- MST 1 ). –MST k = mst(V, E- MST 1 -…-MST k-1 ). MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

10
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

11
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax K-MST K-MST-based graph MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

12
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax K-MST Typical clustering problems –Separated problems and touching problems. –Separated problems includes distance- separated problems and density-separated problems. MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

13
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax K-MST-based clustering Definition of edge weight for separated problems MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

14
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax Three good features: (1) Weights of inter-cluster edges are quite larger than those of intra-cluster edges. (2) The inter- cluster edges are approximately equally distributed to T1 and T2. (3) Except inter- cluster edges, most of edges with large weights come from T2.

15
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

16
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

17
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax K-MST-based clustering Touching problems MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

18
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax Partition(cut1) and Partition(cut3) are similar ; Partition(cut2) and Partition(cut3) are similar.

19
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax Fast approximate MST (FAMST) Traditional MST algorithms take O(N 2 ) time, not favored by large data sets. In practical application, generally FAMST has as same result as exact MST Find a FAMST in O(N 1.55 ) MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

20
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax Fast approximate MST (FAMST) Scheme: Divide-and-Conquer MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

21
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax Fast approximate MST (FAMST) Performance MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

22
University of Joensuu Dept. of Computer Science P.O. Box 111 FIN Joensuu Tel fax MST MST-based clustering K-MST K-MST-based clustering Fast approximate MST

Similar presentations

© 2017 SlidePlayer.com Inc.

All rights reserved.

Ads by Google