國立虎尾科技大學 |

Unsupervised Learning Algorithms for Graph Data.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Unsupervised Learning Algorithms for Graph Data./
Author:	Mohan, Mahesh.
Description:	1 online resource (78 pages)
Notes:	Source: Dissertation Abstracts International, Volume: 79-01(E), Section: B.
Contained By:	Dissertation Abstracts International79-01B(E).
Subject:	Computer science. -
Online resource:	click for full text (PQDT)
ISBN:	9780355205114

Unsupervised Learning Algorithms for Graph Data.
Mohan, Mahesh.

Unsupervised Learning Algorithms for Graph Data. - 1 online resource (78 pages)

Source: Dissertation Abstracts International, Volume: 79-01(E), Section: B.

Thesis (Ph.D.)

Includes bibliographical references

This thesis presents new techniques for unsupervised learning from graph data. Data from many applications, including social networks, transportation systems, images, and climate variables measured over multiple latitudes and longitudes, can be represented as graphs. Graphs provide a convenient representation of relationships between variables. Unsupervised learning is a critical first step in making sense of unlabeled data. The recent surge in the deployment of sensors and data acquisition systems have led to an explosion in the amount of data available, a significant portion of which is unlabeled.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355205114Subjects--Topical Terms:

573171
Computer science.
Index Terms--Genre/Form:

554714
Electronic books.

Unsupervised Learning Algorithms for Graph Data.
LDR:02880ntm a2200373Ki 4500 001 910716
005 20180517112608.5
006 m o u
007 cr mn||||a|a||
008 190606s2017 xx obm 000 0 eng d
020 $a 9780355205114
035 $a (MiAaPQ)AAI10254000
035 $a (MiAaPQ)gwu:13448
035 $a AAI10254000
040 $a MiAaPQ $b eng $c MiAaPQ
099 $a TUL $f hyy $c available through World Wide Web
100 1 $a Mohan, Mahesh. $3 1182151
245 1 0 $a Unsupervised Learning Algorithms for Graph Data.
264 0 $c 2017
300 $a 1 online resource (78 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertation Abstracts International, Volume: 79-01(E), Section: B.
500 $a Adviser: Claire Monteleoni.
502 $a Thesis (Ph.D.) $c The George Washington University $d 2017.
504 $a Includes bibliographical references
520 $a This thesis presents new techniques for unsupervised learning from graph data. Data from many applications, including social networks, transportation systems, images, and climate variables measured over multiple latitudes and longitudes, can be represented as graphs. Graphs provide a convenient representation of relationships between variables. Unsupervised learning is a critical first step in making sense of unlabeled data. The recent surge in the deployment of sensors and data acquisition systems have led to an explosion in the amount of data available, a significant portion of which is unlabeled.
520 $a Many existing approaches for unsupervised machine learning on graphs are computationally expensive and do not scale well to large amounts of data. The work presented here addresses this issue by proposing theoretically justifiable algorithms that provide a good tradeoff between computation time and accuracy. This thesis focuses on two unsupervised learning tasks: spectral clustering and nearest neighbor search using graph kernels.
520 $a I present two novel sampling-based approaches to spectral clustering that scale to a significantly larger number of input points compared to the current state of the art. I also present theoretical results showing that these sampling-based approaches yield a good approximation with respect to a popular spectral clustering objective. In the case of nearest neighbor search, graph kernels are used to propose a hierarchical scheme that allows for efficient queries when the input data has a large number of points in a high dimensional space.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Artificial intelligence. $3 559380
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a The George Washington University. $b Computer Science. $3 1148676
773 0 $t Dissertation Abstracts International $g 79-01B(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10254000 $z click for full text (PQDT)