Outlier detection in astronomical data

Astronomical data sets have experienced an unprecedented and continuing growth in the volume, quality, and complexity over the past few years, driven by the advances in telescope, detector, and computer technology. Like many other fields, astronomy has become a very data rich science. Information content measured in multiple Terabytes, and even larger, multi Petabyte data sets are on the horizon. To cope with this data flood, Virtual Observatory (VO) federates data archives and services representing a new information infrastructure for astronomy of the 21st century and provides the platform to science discovery. Data mining promises to both make the scientific utilization of these data sets more effective and more complete, and to open completely new avenues of astronomical research. Technological problems range from the issues of database design and federation, to data mining and advanced visualization, leading to a new toolkit for astronomical research. This is similar to challenges encountered in other data intensive fields today. Outlier detection is of great importance, as one of four knowledge discovery tasks. The identification of outliers can often lead to the discovery of truly unexpected knowledge in various fields. Especially in astronomy, the great interest of astronomers is to discover unusual, rare or unknown types of astronomical objects or phenomena. The outlier detection approaches in large datasets correctly meet the need of astronomers. In this paper we provide an overview of some techniques for automated identification of outliers in multivariate data. Outliers often provide useful information. Their identification is important not only for improving the analysis but also for indicating anomalies which may require further investigation. The technique may be used in the process of data preprocessing and also be used for preselecting special object candidates.