A Five Step Procedure for Outlier Analysis in Data Mining
Now-a-days, outlier detection is primarily studied as an independent knowledge discovery process merely because outliers might be indicators of interesting events that have never been known before. Despite the advances seen, many issues of outlier detection are left open or not yet completely resolved. Outlier detection is an important data mining task. It deserves more attention from data mining community. There are "Good" outliers that provide useful information that can lead to the discovery of new knowledge and "Bad" outliers that include noisy data points. Distinguishing between different types of outliers is an important issue in many applications. It requires not only an understanding of the mathematical properties of data but also relevant knowledge in the domain context in which the outliers occur.