Homogeneous and heterogeneous distributed classification for pocket data mining

F. Stahl; M. Gaber; P. Aldridge; D. May; Han Liu; Max Bramer; P. Yu

Homogeneous and heterogeneous distributed classification for pocket data mining

F. Stahl, M. Gaber, P. Aldridge, D. May, Han Liu, Max Bramer, P. Yu

School of Computing

Research output: Chapter in Book/Report/Conference proceeding › Chapter (peer-reviewed) › peer-review

142 Downloads (Pure)

Abstract

Pocket Data Mining (PDM) describes the full process of analysing data streams in mobile ad hoc distributed environments. Advances in mobile devices like smart phones and tablet computers have made it possible for a wide range of applications to run in such an environment. In this paper, we propose the adoption of data stream classification techniques for PDM. Evident by a thorough experimental study, it has been proved that running heterogeneous/different, or homogeneous/similar data stream classification techniques over vertically partitioned data (data partitioned according to the feature space) results in comparable performance to batch and centralised learning techniques.

Original language	English
Title of host publication	Transactions on large-scale data and knowledge-centered systems V
Editors	A. Hameurlain, J. Kung, R. Wagner
Place of Publication	Berlin
Publisher	Springer
Pages	183-205
Number of pages	23
Volume	5
Edition	7100
ISBN (Print)	9783642281471
Publication status	Published - 2012

Publication series

Name	Lecture notes in computer sciences
Publisher	Springer
Number	7100

Access to Document

llncs.pdfSubmitted manuscript, 1.15 MB

Cite this

@inbook{c506d578717643b7902c74e7121266ce,

title = "Homogeneous and heterogeneous distributed classification for pocket data mining",

abstract = "Pocket Data Mining (PDM) describes the full process of analysing data streams in mobile ad hoc distributed environments. Advances in mobile devices like smart phones and tablet computers have made it possible for a wide range of applications to run in such an environment. In this paper, we propose the adoption of data stream classification techniques for PDM. Evident by a thorough experimental study, it has been proved that running heterogeneous/different, or homogeneous/similar data stream classification techniques over vertically partitioned data (data partitioned according to the feature space) results in comparable performance to batch and centralised learning techniques.",

author = "F. Stahl and M. Gaber and P. Aldridge and D. May and Han Liu and Max Bramer and P. Yu",

year = "2012",

language = "English",

isbn = "9783642281471",

volume = "5",

series = "Lecture notes in computer sciences",

publisher = "Springer",

number = "7100",

pages = "183--205",

editor = "A. Hameurlain and J. Kung and R. Wagner",

booktitle = "Transactions on large-scale data and knowledge-centered systems V",

edition = "7100",

}

Homogeneous and heterogeneous distributed classification for pocket data mining. / Stahl, F.; Gaber, M.; Aldridge, P. et al.
Transactions on large-scale data and knowledge-centered systems V. ed. / A. Hameurlain; J. Kung; R. Wagner. Vol. 5 7100. ed. Berlin: Springer, 2012. p. 183-205 (Lecture notes in computer sciences; No. 7100).

Research output: Chapter in Book/Report/Conference proceeding › Chapter (peer-reviewed) › peer-review

TY - CHAP

T1 - Homogeneous and heterogeneous distributed classification for pocket data mining

AU - Stahl, F.

AU - Gaber, M.

AU - Aldridge, P.

AU - May, D.

AU - Liu, Han

AU - Bramer, Max

AU - Yu, P.

PY - 2012

Y1 - 2012

N2 - Pocket Data Mining (PDM) describes the full process of analysing data streams in mobile ad hoc distributed environments. Advances in mobile devices like smart phones and tablet computers have made it possible for a wide range of applications to run in such an environment. In this paper, we propose the adoption of data stream classification techniques for PDM. Evident by a thorough experimental study, it has been proved that running heterogeneous/different, or homogeneous/similar data stream classification techniques over vertically partitioned data (data partitioned according to the feature space) results in comparable performance to batch and centralised learning techniques.

AB - Pocket Data Mining (PDM) describes the full process of analysing data streams in mobile ad hoc distributed environments. Advances in mobile devices like smart phones and tablet computers have made it possible for a wide range of applications to run in such an environment. In this paper, we propose the adoption of data stream classification techniques for PDM. Evident by a thorough experimental study, it has been proved that running heterogeneous/different, or homogeneous/similar data stream classification techniques over vertically partitioned data (data partitioned according to the feature space) results in comparable performance to batch and centralised learning techniques.

M3 - Chapter (peer-reviewed)

SN - 9783642281471

VL - 5

T3 - Lecture notes in computer sciences

SP - 183

EP - 205

BT - Transactions on large-scale data and knowledge-centered systems V

A2 - Hameurlain, A.

A2 - Kung, J.

A2 - Wagner, R.

PB - Springer

CY - Berlin

ER -

Homogeneous and heterogeneous distributed classification for pocket data mining

Abstract

Publication series

Access to Document

Fingerprint

Cite this