event thumbnail image
Stability and Resampling Methods for Clustering
Pascal

On a statistical model of cluster stability

author: Zeev Volkovich, ORT Braude College of Engineering
You might be experiencing some problems with Your Video player.
Slides
0:00 ON STATISTICAL MODEL OF CLUSTER STABILITY
0:13 Concept
0:54 Motivating works
1:28 Clustering
1:59 Clustering (cont)
2:26 Example: a three-cluster set partitioned into 2 and 4 clusters
3:35 Implication
3:55 Concept
4:18 Concept (cont. 1)
4:43 Concept (cont. 2)
6:07 Some probability metrics
8:35 Examples
8:49 Ky Fan metrics
9:08 Concentration measure index
10:38 Simple and Compound Metrics
13:38 Geometrical Algorithm
14:06 General algorithm. Given a probability metric dis(·, ·)
16:23 Klebanov’s N-distances
17:25 Simple distances (cont)
18:30 Graphical illustration
20:57 Graphical illustration (cont. 1) Distances between points belonging to different samples
21:22 Graphical illustration (cont. 2)
23:23 Remark
23:56 Euclidean Minimal Spanning Tree
24:49 An EMST of 60 random points
25:01 How can an EMST be used in the cluster validation problem?
25:27 Graphical illustration. Stable clustering
26:21 Graphical illustration. Non-stable clustering
33:58 The two-sample MST-test (cont. 2)
35:09 The two-sample MST-test (cont. 3)
35:36 Theorem’s application
35:59 Theorem’s application (cont. 1)
36:14 Example: Calculation of Rn(S1,S2)
37:38 Distances from normality
38:39 The Kolmogorov-Smirnov Distance
39:24 Example : synthetic data
39:37 Example : synthetic data (cont. 1)
41:42 Membership Stability Algorithm
42:48 A family of clustering algorithms
43:41 Clusters correspondence problem
44:05 Correspondence between labels a and ß obtained for a sample S.
44:43 Example:The Iris Flower Dataset
45:44 Graph of the normalized mean value
45:59 Graph of the normalized quartile value
46:33 Histograms of the distances’ values

Lecture rating

People found this lecture:
Worth seeing
because it is:
 Valuable and informative
Well presented
Easily understandable
Acceptably recorded
You need to login to cast your vote.

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.

Link this page

Would you like to put a link to this lecture on your homepage?
Go ahead! Copy the HTML snippet !

Write your own review or comment: