Mining Approximate Frequent Itemsets over Data Streams

Mining Approximate Frequent Itemsets over Data Streams

This paper proposes a method based on Lossy Counting to mine frequent itemsets. Logarithmic tilted time window is adopted to emphasize the importance of recent data. Multilayer count queue framework is used to avoid the counter overflowing and query top-K itemsets quickly using a index table.

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Applied Mechanics and Materials 2014-10, Vol.685 (Machine, Industry and Manufacturing Based on Applied-Information Technology IV), p.536-539
Hauptverfasser: Yan, Chang Qing, Su, Na, Wu, Zhe Hui, Liu, Ji Min, Liu, Tai An, An, Xin Jun
Format: Artikel
Sprache:eng
Schlagworte:
Approximation
Counting
Data transmission
Multilayers
Query processing
Queues
Tables (data)
Windows (intervals)
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper proposes a method based on Lossy Counting to mine frequent itemsets. Logarithmic tilted time window is adopted to emphasize the importance of recent data. Multilayer count queue framework is used to avoid the counter overflowing and query top-K itemsets quickly using a index table.
ISSN:1660-9336
1662-7482
1662-7482
DOI:10.4028/www.scientific.net/AMM.685.536