10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.
-
date post
21-Dec-2015 -
Category
Documents
-
view
219 -
download
2
Transcript of 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.
![Page 1: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/1.jpg)
10-603/15-826A:Multimedia Databases and Data
Mining
SVD - part I (definitions)
C. Faloutsos
![Page 2: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/2.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 2
Outline
Goal: ‘Find similar / interesting things’
• Intro to DB
• Indexing - similarity search
• Data Mining
![Page 3: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/3.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 3
Indexing - Detailed outline• primary key indexing• secondary key / multi-key indexing• spatial access methods• fractals• text• Singular Value Decomposition (SVD)• multimedia• ...
![Page 4: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/4.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 4
SVD - Detailed outline• Motivation• Definition - properties• Interpretation• Complexity• Case studies• Additional properties
![Page 5: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/5.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 5
SVD - Motivation• problem #1: text - LSI: find ‘concepts’• problem #2: compression / dim. reduction
![Page 6: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/6.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 6
SVD - Motivation• problem #1: text - LSI: find ‘concepts’
![Page 7: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/7.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 7
SVD - Motivation• problem #2: compress / reduce
dimensionality
![Page 8: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/8.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 8
Problem - specs
• ~10**6 rows; ~10**3 columns; no updates;
• random access to any cell(s) ; small error: OK
![Page 9: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/9.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 9
SVD - Motivation
![Page 10: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/10.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 10
SVD - Motivation
![Page 11: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/11.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 11
SVD - Detailed outline• Motivation• Definition - properties• Interpretation• Complexity• Case studies• Additional properties
![Page 12: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/12.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 12
SVD - Definition(reminder: matrix multiplication
1 2
3 45 6
x 1
-1
3 x 2 2 x 1
=
![Page 13: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/13.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 13
SVD - Definition(reminder: matrix multiplication
1 2
3 45 6
x 1
-1
3 x 2 2 x 1
=
3 x 1
![Page 14: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/14.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 14
SVD - Definition(reminder: matrix multiplication
1 2
3 45 6
x 1
-1
3 x 2 2 x 1
=
-1
3 x 1
![Page 15: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/15.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 15
SVD - Definition(reminder: matrix multiplication
1 2
3 45 6
x 1
-1
3 x 2 2 x 1
=
-1
-1
3 x 1
![Page 16: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/16.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 16
SVD - Definition(reminder: matrix multiplication
1 2
3 45 6
x 1
-1=
-1
-1-1
![Page 17: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/17.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 17
SVD - Definition
A[n x m] = U[n x r] r x r] (V[m x r])T
• A: n x m matrix (eg., n documents, m terms)
• U: n x r matrix (n documents, r concepts)• : r x r diagonal matrix (strength of each
‘concept’) (r : rank of the matrix)• V: m x r matrix (m terms, r concepts)
![Page 18: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/18.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 18
SVD - Definition• A = U VT - example:
![Page 19: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/19.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 19
SVD - PropertiesTHEOREM [Press+92]: always possible to
decompose matrix A into A = U VT , where• U, V: unique (*)• U, V: column orthonormal (ie., columns are
unit vectors, orthogonal to each other)– UT U = I; VT V = I (I: identity matrix)
• : eigenvalues are positive, and sorted in decreasing order
![Page 20: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/20.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 20
SVD - Example• A = U VT - example:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
datainf.
retrieval
brain lung
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=CS
MD
9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
![Page 21: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/21.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 21
SVD - Example• A = U VT - example:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
datainf.
retrieval
brain lung
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=CS
MD
9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
CS-conceptMD-concept
![Page 22: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/22.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 22
SVD - Example• A = U VT - example:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
datainf.
retrieval
brain lung
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=CS
MD
9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
CS-conceptMD-concept
doc-to-concept similarity matrix
![Page 23: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/23.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 23
SVD - Example• A = U VT - example:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
datainf.
retrieval
brain lung
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=CS
MD
9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
‘strength’ of CS-concept
![Page 24: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/24.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 24
SVD - Example• A = U VT - example:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
datainf.
retrieval
brain lung
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=CS
MD
9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
term-to-conceptsimilarity matrix
CS-concept
![Page 25: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/25.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 25
SVD - Example• A = U VT - example:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
datainf.
retrieval
brain lung
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=CS
MD
9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
term-to-conceptsimilarity matrix
CS-concept
![Page 26: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/26.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 26
SVD - Detailed outline• Motivation• Definition - properties• Interpretation• Complexity• Case studies• Additional properties
![Page 27: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/27.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 27
SVD - Interpretation #1‘documents’, ‘terms’ and ‘concepts’:• U: document-to-concept similarity matrix• V: term-to-concept sim. matrix• : its diagonal elements: ‘strength’ of each
concept
![Page 28: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/28.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 28
SVD - Interpretation #2• best axis to project on: (‘best’ = min sum of
squares of projection errors)
![Page 29: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/29.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 29
SVD - Motivation
![Page 30: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/30.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 30
SVD - interpretation #2
• minimum RMS error
SVD: givesbest axis to project
v1
![Page 31: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/31.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 31
SVD - Interpretation #2
![Page 32: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/32.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 32
SVD - Interpretation #2• A = U VT - example:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
v1
![Page 33: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/33.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 33
SVD - Interpretation #2• A = U VT - example:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
variance (‘spread’) on the v1 axis
![Page 34: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/34.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 34
SVD - Interpretation #2• A = U VT - example:
– U gives the coordinates of the points in the projection axis
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
![Page 35: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/35.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 35
SVD - Interpretation #2• More details• Q: how exactly is dim. reduction done?
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
![Page 36: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/36.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 36
SVD - Interpretation #2• More details• Q: how exactly is dim. reduction done?• A: set the smallest eigenvalues to zero:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
![Page 37: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/37.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 37
SVD - Interpretation #2
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
~9.64 0
0 0x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
![Page 38: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/38.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 38
SVD - Interpretation #2
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
~9.64 0
0 0x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
![Page 39: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/39.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 39
SVD - Interpretation #2
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
0.18
0.36
0.18
0.90
0
00
~9.64
x
0.58 0.58 0.58 0 0
x
![Page 40: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/40.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 40
SVD - Interpretation #2
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
~
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 0 0
0 0 0 0 00 0 0 0 0
![Page 41: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/41.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 41
SVD - Interpretation #2Exactly equivalent:‘spectral decomposition’ of the matrix:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
![Page 42: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/42.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 42
SVD - Interpretation #2Exactly equivalent:‘spectral decomposition’ of the matrix:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
= x xu1 u2
1
2
v1
v2
![Page 43: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/43.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 43
SVD - Interpretation #2Exactly equivalent:‘spectral decomposition’ of the matrix:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
= u11 vT1 u22 vT
2+ +...n
m
![Page 44: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/44.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 44
SVD - Interpretation #2Exactly equivalent:‘spectral decomposition’ of the matrix:
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
= u11 vT1 u22 vT
2+ +...n
m
n x 1 1 x m
r terms
![Page 45: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/45.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 45
SVD - Interpretation #2approximation / dim. reduction:by keeping the first few terms (Q: how many?)
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
= u11 vT1 u22 vT
2+ +...n
m
assume: 1 >= 2 >= ...
![Page 46: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/46.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 46
SVD - Interpretation #2A (heuristic - [Fukunaga]): keep 80-90% of
‘energy’ (= sum of squares of i ’s)
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
= u11 vT1 u22 vT
2+ +...n
m
assume: 1 >= 2 >= ...
![Page 47: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/47.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 47
SVD - Detailed outline• Motivation• Definition - properties• Interpretation
– #1: documents/terms/concepts– #2: dim. reduction– #3: picking non-zero, rectangular ‘blobs’
• Complexity• Case studies• Additional properties
![Page 48: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/48.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 48
SVD - Interpretation #3• finds non-zero ‘blobs’ in a data matrix
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
![Page 49: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/49.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 49
SVD - Interpretation #3• finds non-zero ‘blobs’ in a data matrix
1 1 1 0 0
2 2 2 0 0
1 1 1 0 0
5 5 5 0 0
0 0 0 2 2
0 0 0 3 30 0 0 1 1
0.18 0
0.36 0
0.18 0
0.90 0
0 0.53
0 0.800 0.27
=9.64 0
0 5.29x
0.58 0.58 0.58 0 0
0 0 0 0.71 0.71
x
![Page 50: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/50.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 50
SVD - Interpretation #3• Drill: find the SVD, ‘by inspection’!• Q: rank = ??
1 1 1 0 0
1 1 1 0 0
1 1 1 0 0
0 0 0 1 1
0 0 0 1 1
= x x?? ??
??
![Page 51: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/51.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 51
SVD - Interpretation #3• A: rank = 2 (2 linearly independent
rows/cols)
1 1 1 0 0
1 1 1 0 0
1 1 1 0 0
0 0 0 1 1
0 0 0 1 1
= x x??
??
?? 0
0 ??
??
??
![Page 52: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/52.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 52
SVD - Interpretation #3• A: rank = 2 (2 linearly independent
rows/cols)
1 1 1 0 0
1 1 1 0 0
1 1 1 0 0
0 0 0 1 1
0 0 0 1 1
= x x?? 0
0 ??
1 0
1 0
1 0
0 1
0 11 1 1 0 0
0 0 0 1 1
orthogonal??
![Page 53: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/53.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 53
SVD - Interpretation #3• column vectors: are orthogonal - but not
unit vectors:
1 1 1 0 0
1 1 1 0 0
1 1 1 0 0
0 0 0 1 1
0 0 0 1 1
= x x?? 0
0 ??
1/sqrt(3) 0
1/sqrt(3) 0
1/sqrt(3) 0
0 1/sqrt(2)
0 1/sqrt(2)
1/sqrt(3) 1/sqrt(3) 1/sqrt(3) 0 0
0 0 0 1/sqrt(2) 1/sqrt(2)
![Page 54: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/54.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 54
SVD - Interpretation #3• and the eigenvalues are:
1 1 1 0 0
1 1 1 0 0
1 1 1 0 0
0 0 0 1 1
0 0 0 1 1
= x x3 0
0 2
1/sqrt(3) 0
1/sqrt(3) 0
1/sqrt(3) 0
0 1/sqrt(2)
0 1/sqrt(2)
1/sqrt(3) 1/sqrt(3) 1/sqrt(3) 0 0
0 0 0 1/sqrt(2) 1/sqrt(2)
![Page 55: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/55.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 55
SVD - Interpretation #3• Q: How to check we are correct?
1 1 1 0 0
1 1 1 0 0
1 1 1 0 0
0 0 0 1 1
0 0 0 1 1
= x x3 0
0 2
1/sqrt(3) 0
1/sqrt(3) 0
1/sqrt(3) 0
0 1/sqrt(2)
0 1/sqrt(2)
1/sqrt(3) 1/sqrt(3) 1/sqrt(3) 0 0
0 0 0 1/sqrt(2) 1/sqrt(2)
![Page 56: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/56.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 56
SVD - Interpretation #3• A: SVD properties:
– matrix product should give back matrix A– matrix U should be column-orthonormal, i.e.,
columns should be unit vectors, orthogonal to each other
– ditto for matrix V– matrix should be diagonal, with positive
values
![Page 57: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/57.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 57
SVD - Detailed outline• Motivation• Definition - properties• Interpretation• Complexity• Case studies• Additional properties
![Page 58: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/58.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 58
SVD - Complexity• O( n * m * m) or O( n * n * m) (whichever
is less)• less work, if we just want eigenvalues• or if we want first k eigenvectors• or if the matrix is sparse [Berry]• Implemented: in any linear algebra package
(LINPACK, matlab, Splus, mathematica ...)
![Page 59: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/59.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 59
SVD - conclusions so far• SVD: A= U VT : unique (*)• U: document-to-concept similarities• V: term-to-concept similarities• : strength of each concept• dim. reduction: keep the first few strongest
eigenvalues (80-90% of ‘energy’)– SVD: picks up linear correlations
• SVD: picks up non-zero ‘blobs’
![Page 60: 10-603/15-826A: Multimedia Databases and Data Mining SVD - part I (definitions) C. Faloutsos.](https://reader035.fdocuments.net/reader035/viewer/2022062516/56649d615503460f94a421ab/html5/thumbnails/60.jpg)
Multi DB and D.M. Copyright: C. Faloutsos (2001) 60
References• Berry, Michael: http://www.cs.utk.edu/~lsi/
• Fukunaga, K. (1990). Introduction to Statistical Pattern Recognition, Academic Press.
• Press, W. H., S. A. Teukolsky, et al. (1992). Numerical Recipes in C, Cambridge University Press.