國立高雄大學圖資館 |

Language: English

Back

Data Mining Compressed, Incomplete a...

Hunter, Blake.

Data Mining Compressed, Incomplete and Inaccurate High Dimensional Data.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Data Mining Compressed, Incomplete and Inaccurate High Dimensional Data.
Author:	Hunter, Blake.
Description:	99 p.
Notes:	Source: Dissertation Abstracts International, Volume: 73-01, Section: B, page: 0402.
Notes:	Adviser: Thomas Strohmer.
Contained By:	Dissertation Abstracts International73-01B.
Subject:	Applied Mathematics.
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3474399
ISBN:	9781124907291

Data Mining Compressed, Incomplete and Inaccurate High Dimensional Data.
Hunter, Blake.

Data Mining Compressed, Incomplete and Inaccurate High Dimensional Data. - 99 p.

Source: Dissertation Abstracts International, Volume: 73-01, Section: B, page: 0402.

Thesis (Ph.D.)--University of California, Davis, 2011.

As the size and complexity of data continues to grow, extracting knowledge becomes exponentially more challenging. Active areas of research for mining this high dimensional data can be found across a broad range of scientific fields including pure and applied mathematics, statistics, computer science and engineering. Spectral embedding is one of the most powerful and widely used techniques for extracting the underlying global structure of a data set. Compressed sensing and matrix completion have emerged as prevailing methods for efficiently recovering sparse and partially observed signals. In this dissertation, the distance preserving measurements of compressed sensing and matrix completion are combined with the robust power of spectral embedding. The analysis provides rigorous bounds on how small perturbations from using compressed sensing and matrix completion affect the affinity matrix and in succession the spectral coordinates.

ISBN: 9781124907291Subjects--Topical Terms:

530992
Applied Mathematics.

Data Mining Compressed, Incomplete and Inaccurate High Dimensional Data.
LDR:01953nmm 2200313 4500 001 380629
005 20130530092705.5
008 130708s2011 ||||||||||||||||| ||eng d
020 $a 9781124907291
035 $a (UMI)AAI3474399
035 $a AAI3474399
040 $a UMI $c UMI
100 1 $a Hunter, Blake. $3 603201
245 1 0 $a Data Mining Compressed, Incomplete and Inaccurate High Dimensional Data.
300 $a 99 p.
500 $a Source: Dissertation Abstracts International, Volume: 73-01, Section: B, page: 0402.
500 $a Adviser: Thomas Strohmer.
502 $a Thesis (Ph.D.)--University of California, Davis, 2011.
520 $a As the size and complexity of data continues to grow, extracting knowledge becomes exponentially more challenging. Active areas of research for mining this high dimensional data can be found across a broad range of scientific fields including pure and applied mathematics, statistics, computer science and engineering. Spectral embedding is one of the most powerful and widely used techniques for extracting the underlying global structure of a data set. Compressed sensing and matrix completion have emerged as prevailing methods for efficiently recovering sparse and partially observed signals. In this dissertation, the distance preserving measurements of compressed sensing and matrix completion are combined with the robust power of spectral embedding. The analysis provides rigorous bounds on how small perturbations from using compressed sensing and matrix completion affect the affinity matrix and in succession the spectral coordinates.
590 $a School code: 0029.
650 4 $a Applied Mathematics. $3 530992
650 4 $a Mathematics. $3 184409
650 4 $a Statistics. $3 182057
690 $a 0364
690 $a 0405
690 $a 0463
710 2 $a University of California, Davis. $b Mathematics. $3 603202
773 0 $t Dissertation Abstracts International $g 73-01B.
790 1 0 $a Strohmer, Thomas, $e advisor
790 1 0 $a Saito, Naoki $e committee member
790 1 0 $a Bremer, James $e committee member
790 $a 0029
791 $a Ph.D.
792 $a 2011
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3474399