Speed and Size Characteristic of Concept Drift

Warning

This publication doesn't include Institute of Computer Science. It includes Faculty of Informatics. Official publication website can be found on muni.cz.
Authors

KOSINA Petr GAMA Joao SEBASTIAO Raquel

Year of publication 2010
Type Article in Proceedings
MU Faculty or unit

Faculty of Informatics

Citation
Web http://www.liaad.up.pt/udm/procUDMECAI2010.pdf
Keywords Data Streams; Concept Drift; Drift Rate
Description Stream mining drags more and more attention in data mining since the volume of available data becomes too large and problematic for conventional mining tools. One typical feature of data stream is a change over time in the underlying data distribution. This feature is particularly interesting for studying, and vital for quick adaptation to the new situation. The focus of this paper is in small detail in the process of drift detection - the connection between the area affected by the change and number of errors. The goal is to provide the information about the size and speed characteristics of the change to the user.
Related projects:

You are running an old browser version. We recommend updating your browser to its latest version.

More info