Dependency detection in MobiMine: A systems perspective

Sweta Pittie; Hillol Kargupta; Byung Hoon Park

doi:10.1016/S0020-0255(03)00171-3

Dependency detection in MobiMine: A systems perspective

Sweta Pittie, Hillol Kargupta, Byung Hoon Park

Research output: Contribution to journal › Article › peer-review

12 Scopus citations

Abstract

This paper considers the problem of detecting dependencies among data streams and presenting the results in a mobile data mining system. It particularly focuses on the systems issues addressed by MobiMine, a system for mining financial data streams from PDAs. It presents an overview of the MobiMine, explains the two algorithmic techniques (correlation and conditional probability rules) used for detecting dependencies between a pair of stocks, identifies the systems challenges, and offers solutions. The paper also presents experimental results supporting MobiMine's scalable performance.

Original language	English
Pages (from-to)	227-243
Number of pages	17
Journal	Information Sciences
Volume	155
Issue number	3-4
DOIs	https://doi.org/10.1016/S0020-0255(03)00171-3
State	Published - Oct 15 2003
Externally published	Yes

Funding

The authors acknowledge supports from the United States National Science Foundation CAREER award IIS-0093353 and TEDCO, Maryland Technology Development Center. The authors would like to thank Patrick Blair for his help in developing the system.

Keywords

Data streams
Dependency detection
Mobile data mining

Access to Document

10.1016/S0020-0255(03)00171-3

Cite this

@article{caa577883056405cb25c30b7924abade,

title = "Dependency detection in MobiMine: A systems perspective",

abstract = "This paper considers the problem of detecting dependencies among data streams and presenting the results in a mobile data mining system. It particularly focuses on the systems issues addressed by MobiMine, a system for mining financial data streams from PDAs. It presents an overview of the MobiMine, explains the two algorithmic techniques (correlation and conditional probability rules) used for detecting dependencies between a pair of stocks, identifies the systems challenges, and offers solutions. The paper also presents experimental results supporting MobiMine's scalable performance.",

keywords = "Data streams, Dependency detection, Mobile data mining",

author = "Sweta Pittie and Hillol Kargupta and Park, {Byung Hoon}",

year = "2003",

month = oct,

day = "15",

doi = "10.1016/S0020-0255(03)00171-3",

language = "English",

volume = "155",

pages = "227--243",

journal = "Information Sciences",

issn = "0020-0255",

publisher = "Elsevier",

number = "3-4",

}

TY - JOUR

T1 - Dependency detection in MobiMine

T2 - A systems perspective

AU - Pittie, Sweta

AU - Kargupta, Hillol

AU - Park, Byung Hoon

PY - 2003/10/15

Y1 - 2003/10/15

N2 - This paper considers the problem of detecting dependencies among data streams and presenting the results in a mobile data mining system. It particularly focuses on the systems issues addressed by MobiMine, a system for mining financial data streams from PDAs. It presents an overview of the MobiMine, explains the two algorithmic techniques (correlation and conditional probability rules) used for detecting dependencies between a pair of stocks, identifies the systems challenges, and offers solutions. The paper also presents experimental results supporting MobiMine's scalable performance.

AB - This paper considers the problem of detecting dependencies among data streams and presenting the results in a mobile data mining system. It particularly focuses on the systems issues addressed by MobiMine, a system for mining financial data streams from PDAs. It presents an overview of the MobiMine, explains the two algorithmic techniques (correlation and conditional probability rules) used for detecting dependencies between a pair of stocks, identifies the systems challenges, and offers solutions. The paper also presents experimental results supporting MobiMine's scalable performance.

KW - Data streams

KW - Dependency detection

KW - Mobile data mining

UR - http://www.scopus.com/inward/record.url?scp=0141457956&partnerID=8YFLogxK

U2 - 10.1016/S0020-0255(03)00171-3

DO - 10.1016/S0020-0255(03)00171-3

M3 - Article

AN - SCOPUS:0141457956

SN - 0020-0255

VL - 155

SP - 227

EP - 243

JO - Information Sciences

JF - Information Sciences

IS - 3-4

ER -

Dependency detection in MobiMine: A systems perspective

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this