| Summary: | This paper is concerned with two methods, one based on eigenvalue analysis, and the other, a modified version of singular value decomposition (SVD) called pseudo-SVD, for detecting outliers in high-dimensional data sets. The eigenvalue analysis approach examines the spatial relationship among the column vectors of object-attribute matrix to obtain an insight into the degree of inconsistency in a cluster of data. The pseudo-SVD method, in which the singular values are allowed to have a sign, looks at the direction of vectors in the object-attribute matrix and based on the degree of their orthogonality detects the outliers. The pseudo-SVD algorithm is formulated as an optimisation problem for clustering the data on the basis of their angular inclination. The methods have been applied to two case studies: one pertaining to a dermatological dataset and the other related to an engineering problem of state estimation. Further research directions are also discussed.
|
|---|
