• DocumentCode
    2348018
  • Title

    Data mining a high-speed bursty stream on a limited buffer in pseudo-stationary states

  • Author

    Sengupta, Sam ; Andriamanalimanana, Bruno ; Card, Stuart W. ; Das, Kaustav ; Sharma, Rachid ; Gunasekaran, Anitha

  • Author_Institution
    Inst. of Technol., State Univ. of New York, Utica, NY
  • fYear
    2003
  • fDate
    8-10 Sept. 2003
  • Firstpage
    201
  • Lastpage
    204
  • Abstract
    Mining a high speed bursty data stream is always a challenge on a limited size buffer. Often a relatively cheaper AMS (anytime mining solution) approach may be the only plausible scheme one could rely on at times. Mining task becomes enormously complicated when the first-level buffer has to host several dependent streams. This becomes worse when incoming data streams take time to settle down in their respective steady states. A buffer sharing and capture models are indicated for some simple situations involving multiple streams. These models could be extended to generalize a linear buffer model to a hierarchical model
  • Keywords
    buffer storage; data mining; AMS; anytime mining solution; buffer acquisition; buffer sharing; data mining; hierarchical buffer model; multiple data stream mining; Buffer storage; Data mining; Databases; Internet; Knowledge engineering; Sampling methods; Stationary state;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, 2003. Proceedings of the Second IEEE International Workshop on
  • Conference_Location
    Lviv
  • Print_ISBN
    0-7803-8138-6
  • Type

    conf

  • DOI
    10.1109/IDAACS.2003.1249549
  • Filename
    1249549