Homogeneous and heterogeneous distributed classification for pocket data mining

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

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

84 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 languageEnglish
Title of host publicationTransactions on large-scale data and knowledge-centered systems V
EditorsA. Hameurlain, J. Kung, R. Wagner
Place of PublicationBerlin
PublisherSpringer
Pages183-205
Number of pages23
Volume5
Edition7100
ISBN (Print)9783642281471
Publication statusPublished - 2012

Publication series

NameLecture notes in computer sciences
PublisherSpringer
Number7100

Fingerprint

Dive into the research topics of 'Homogeneous and heterogeneous distributed classification for pocket data mining'. Together they form a unique fingerprint.

Cite this