Detecting Outliers on Arbitrary Data Streams using Anytime Approaches

Data streams are increasingly gaining importance in many sensoring and monitoring environments. Frequent mining tasks on data streams include classification, modeling and outlier detection. Since often the data arrival rates vary, anytime algorithms have been proposed for stream clustering and classification, which can deliver a fast first result and improve their result if more time is available.

In this work, we propose the novel concept of anytime outlier detection and introduce an algorithm for anytime outlier detection based on a hierarchical cluster representation. We show promising results in preliminary experiments and discuss future research directions for anytime outlier detection.

Authors: Assent I., Kranen P., Baldauf C., Seidl T.
Published in: International Workshop on Novel Data Stream Pattern Mining Techniques (StreamKDD 2010) in conjunction with 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD 2010), Washington, DC, USA
Publisher: ACM - New York, NY, USA
Sprache: EN
Jahr: 2010
Seiten: 10-16
ISBN: 978-1-4503-0226-5
Konferenz: StreamKDD @KDD
DOI:10.1145/1833280.1833282
URL:StreamKDD 2010
online proceedings
Typ: Tagungsbeiträge
Forschungsgebiet: Data Analysis and Knowledge Extraction