Abstract
We consider a CPU constrained environment for finding approximation of frequent sets in data streams using the landmark window. Our algorithm can detect overload situations, i.e., breaching the CPU capacity, and sheds data in the stream to "keep up". This is done within a controlled error threshold by exploiting the Chernoff-bound. Empirical evaluation of the algorithm confirms the feasibility. © 2012 Springer-Verlag.
Cite
CITATION STYLE
Dang, X. H., Ong, K. L., & Lee, V. (2012). An adaptive algorithm for finding frequent sets in landmark windows. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7520 LNAI, pp. 590–597). https://doi.org/10.1007/978-3-642-33362-0_47
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
