International Research Journal of Education and Technology

Data Mining is the process of finding correlations or hidden patterns among the dozens of fields in a large database. There are many techniques available in data mining to find relations, among them clustering and classification are the two common techniques for finding hidden patterns. Here, we considered clustering which is most important technique for exploring data mining. Clustering is the process of grouping the related records based on their similarity and dissimilarity. In clustering, outlier is one of the major issue, that reduce accuracy of data. Outlier is an observation that is in an abnormal distance from other values in the dataset. Outlier occurs due to the changes that caused in system behavior, human error, poor data quality or by fraudulent behavior. This paper deals with the study of outliers, their types, applications and hierarchical clustering algorithms that deals with outlier detection.

Key Words:  Outliers, Approaches, Applications, CURE, BIRCH, ROCK, CHAMELEON