0.25
0.5
0.75
1.25
1.5
1.75
2
Clustering Rankings in the Fourier Domain
Published on Oct 03, 20112564 Views
It is the purpose of this paper to introduce a novel approach to clustering rank data on a set of possibly large cardinality n ∈ ℕ*, relying upon Fourier representation of functions defined on the sym
Related categories
Chapter list
Clustering Rankings in the Fourier Domain00:00
Distributions on rankings00:27
Representation for distributions on Rankings (1)01:47
Representation for distributions on Rankings (2)02:18
Contributions02:40
Outline03:12
Fourier representation (1)03:29
Fourier representation (2)03:48
Fourier representation (3)04:22
Example: Mallows05:46
Uncertainty principle06:42
Outline: Sparse Clustering of Rankings08:35
Clustering of rankings08:47
Managing sparsity09:25
Algorithm10:01
Outline: Numerical Experiments11:22
Experiments11:26
Mallows11:52
Top-4 lists12:36
E-commerce dataset13:03
Conclusion and perspectives13:39
Thank you14:32
Bibliography14:38