Clustered and Optimized Adaptive Antihub Technique Coaat for Unsupervised Outlier Detection

Abstract

Recently, there are voluminous application domains where the data is of considerably higher dimensionality describing the facts via accumulation of attributes. Moreover outlier detection which aims to find objects that are considerably dissimilar, exceptional and inconsistent with respect to the majority of data in an input data base, becomes an emerging research area in data mining since it acts as an important data preprocessing and cleaning technique to remove outlier objects. Recognition of outlier objects in dataset is useful and can be demonstrated to make the condition of results better before data mining applications make out its processing straight away. newline