An Efficient Partitioning Algorithm to Find Un-Expected Behavioural Data Points
In Data Mining an outlier is an exception that deviates much from other observations in the multidimensional space. There are various approaches to detect outliers in the data set. Many different approaches are proposed by the researchers. In this paper three distance based outlier detection algorithms are taken and compared with the proposed algorithm. Outlier detection is to the problem of finding patterns in data that do not match to expected behaviour. Outlier detection finds extensive use in a wide range of applications such as fraud detection in credit card usage, insurance or health care, intrusion detection in cyber security, fault detection in safety critical systems and military surveillance of enemy activities.