중심도와 중심 근접도를 이용한 효과적인 아웃라이어 검출 방법

초록

An outlier in data is an observation or an object that is considerably dissimilar or inconsistent with the remainder of the data. In this paper, we first propose the concept of Centrality and Center Proximity which can consider the characteristics of all of the objects in the data set. We, then, propose a novel graph-based outlier detection method which can solve the problems of local density, micro cluster, and fringe objects. Finally, we conduct extensive experiments with various datasets and show the effectiveness of the proposed method.

키워드