A Novel Geometric Approach for Outlier Recognition in High Dimension

TL;DR

This paper introduces a new geometric algorithm for high-dimensional outlier detection that is efficient, scalable, and adaptable to multi-class inliers, outperforming existing methods in accuracy and speed.

Contribution

The paper presents a novel core-set based algorithm for outlier recognition that operates in linear time and space and extends naturally to multi-class inliers.

Findings

Outperforms existing algorithms on benchmark datasets

Operates with linear time and space complexity

Provides theoretical guarantees of quality

Abstract

Outlier recognition is a fundamental problem in data analysis and has attracted a great deal of attention in the past decades. However, most existing methods still suffer from several issues such as high time and space complexities or unstable performances for different datasets. In this paper, we provide a novel algorithm for outlier recognition in high dimension based on the elegant geometric technique ``core-set". The algorithm needs only linear time and space complexities and achieves a solid theoretical quality guarantee. Another advantage over the existing methods is that our algorithm can be naturally extended to handle multi-class inliers. Our experimental results show that our algorithm outperforms existing algorithms on both random and benchmark datasets.