Subspace outlier mining in large multimedia databases

Parallel Universes and Local Patterns

计算机科学;物理学

Ira Assent ; Ralph Krieger ; Emmanuel Müller ; Thomas Seidl

Others  :  http://drops.dagstuhl.de/opus/volltexte/2007/1257/pdf/07181.AssentIra.ExtAbstract.1257.pdf
PID  :  10578

学科分类：计算机科学（综合）

来源: CEUR

【 摘 要 】

Increasingly large multimedia databases in life sciences, e-commerce, or monitoring applications cannot be browsed manually, but require automatic knowledge discovery in databases (KDD) techniques to detect novel and interesting patterns. Clustering, aims at grouping similar objects into clusters, separating dissimilar objects. Density-based clustering has been shown to detect arbitrarily shaped clusters even innoisy data bases. In high-dimensional data bases, meaningful clusters can no longer be detected due to the curse of dimensionality. Consequently,subspace clustering searches for clusters hidden in any subset of the set of dimensions. Clustering information is very useful for applications like fraud detection where outliers, i.e. objects which differ from all clusters, are searched. We propose a density-based subspace clustering model for outlier detection. We define outliers with respect to maximal and non-redundant subspace clusters. We demonstrate the quality of our subspace clustering results in experiments on real world databases and discuss our outlier model as well as future work.