Outlier Detection Algorithms in Data Mining

Abstract

Outlier is defined as an observation that deviates too much from other observations. The identification of outliers can lead to the discovery of useful and meaningful knowledge. Outlier detection has been extensively studied in the past decades. However, most existing research focuses on the algorithm based on special background, compared with outlier detection approach is still rare. This paper mainly discusses and compares approach of different outlier detection from data mining perspective, which can be categorized into two categories: classic outlier approach and spatial outlier approach. The classic outlier approach analyzes outlier based on transaction dataset, which can be grouped into statistical-based approach, distance-based approach, deviation-based approach, density-based approach. The spatial outlier approach analyzes outlier based on spatial dataset that non-spatial and spatial data are significantly different from transaction data, which can be grouped into space-based approach and graph-based approach. Finally, the paper concludes some advances in outlier detection recently.