0.25
0.5
0.75
1.25
1.5
1.75
2
The Minimum Code Length for Clustering Using the Gray Code
Published on Oct 03, 20112516 Views
We propose new approaches to exploit compression algorithms for clustering numerical data. Our first contribution is to design a measure that can score the quality of a given clustering result under t
Related categories
Chapter list
The Minimum Code Length for Clustering Using the Gray Code00:00
Contributions00:01
Demonstration (Synthetic Dataset)01:07
G-COOL01:17
K-means01:32
Results (Real datasets) - 101:43
Results (Real datasets) - 202:01
Outline (1)02:34
Outline (2)02:36
Clustering Focusing on Compression02:40
Our Strategy (1)03:56
Our Strategy (2)04:20
Outline (3)05:35
MCL (Minimum Code Length) - 105:46
MCL (Minimum Code Length) - 206:05
Binary Encoding06:37
MCL with Binary Encoding (1)07:10
MCL with Binary Encoding (2)07:19
MCL with Binary Encoding (3)07:24
MCL with Binary Encoding (4)07:34
MCL with Binary Encoding (5)07:44
MCL with Binary Encoding (6)08:02
MCL with Binary Encoding (7)08:07
MCL with Binary Encoding (8)08:14
MCL with Binary Encoding (9)08:19
MCL with Binary Encoding (10)08:24
Definition of MCL08:47
Minimizing MCL and Clustering08:53
Outline (4)09:21
Optimization by COOL09:28
COOL with Binary Encoding (1)09:58
COOL with Binary Encoding (2)10:12
COOL with Binary Encoding (3)10:19
COOL with Binary Encoding (4)10:36
COOL with Binary Encoding (5)11:02
COOL with Binary Encoding (6)11:09
COOL with Binary Encoding (7)11:21
COOL with Binary Encoding (8)11:24
Noise Filtering by COOL (1)11:30
Noise Filtering by COOL (2)11:51
Algorithm of COOL11:57
Outline (5)12:01
Gray Code12:07
Gray Code Embedding12:53
COOL with Gray Code (G-COOL) - 113:23
COOL with Gray Code (G-COOL) - 213:30
COOL with Gray Code (G-COOL) - 313:45
COOL with Gray Code (G-COOL) - 414:04
COOL with Gray Code (G-COOL) - 514:12
COOL with Gray Code (G-COOL) - 614:17
COOL with Gray Code (G-COOL) - 714:20
COOL with Gray Code (G-COOL) - 814:23
COOL with Binary Encoding (9)14:30
Theoretical Analysis of G-COOL14:42
Demonstration of G-COOL15:21
Outline (6)15:37
Experimental Methods15:39
Results (Synthetic datasets) (1)15:56
Results (Synthetic datasets) (2)16:13
Results (Synthetic datasets) (3)16:23
Results (Synthetic datasets) (4)16:56
Results (Synthetic datasets) (5)17:10
Results (Synthetic datasets) (6)17:20
Results (Real datasets) (1)17:29
Results (Real datasets) (2)17:33
Results (Real datasets) (3)17:38
Outline (7)17:39
Conclusion17:41