Abstract/Details

PADMINI: A peer-to-peer distributed data mining system for astronomy researchers


2010 2010

Other formats: Order a copy

Abstract (summary)

As the amount of data available at geographically distributed sources increases rapidly, the need for efficient distributed data mining is becoming increasingly important. Increasing computation powers (change this) at lower hardware costs and reliable communication mechanisms have also led to the proliferation of Peer-to-Peer networks. These factors have lead to the development of dedicated distributed solutions that can run on Peer-to-Peer networks. Many domains such as finance, astronomy, bioinformatics etc. face varied challenges where such solutions can prove instrumental. This thesis presents PADMINI—a Peer-to-Peer Astronomy Data Mining system. Unlike centralized data mining systems, PADMINI is a Web based system powered by Google Sky and distributed data mining algorithms that run on a collection of computing nodes. PADMINI supports two disparate frameworks, namely Hadoop and Distributed Data Mining Toolkit. These frameworks enable PADMINI to support a wide range of data mining algorithms. This work presents solutions implemented on PADMINI for specific data mining problems like Outlier Detection and Classifier Learning. The PADMINI system can also be used to learn (classifiers) classification models from any source of data over the internet, without requiring any kind of support from the host servers. Experimental results to establish the correctness of the solutions and the scalable nature of the PADMINI system are also provided.

Indexing (details)


Subject
Computer science
Classification
0984: Computer science
Identifier / keyword
Applied sciences; Data mining; Distributed; P2p
Title
PADMINI: A peer-to-peer distributed data mining system for astronomy researchers
Author
Mahule, Tushar Pradeep
Number of pages
112
Publication year
2010
Degree date
2010
School code
0434
Source
MAI 49/01M, Masters Abstracts International
Place of publication
Ann Arbor
Country of publication
United States
ISBN
9781124226958
Advisor
Kargupta, Hillol
Committee member
Borne, Kirk; Oates, Tim; Peng, Yun
University/institution
University of Maryland, Baltimore County
Department
Computer Science
University location
United States -- Maryland
Degree
M.S.
Source type
Dissertations & Theses
Language
English
Document type
Dissertation/Thesis
Dissertation/thesis number
1481234
ProQuest document ID
757888801
Copyright
Database copyright ProQuest LLC; ProQuest does not claim copyright in the individual underlying works.
Document URL
http://search.proquest.com/docview/757888801
Access the complete full text

You can get the full text of this document if it is part of your institution's ProQuest subscription.

Try one of the following:

  • Connect to ProQuest through your library network and search for the document from there.
  • Request the document from your library.
  • Go to the ProQuest login page and enter a ProQuest or My Research username / password.