Outlier Detection Using Enhanced K-means Clustering Algorithm and Weight Based Center Approach

ABSTRACT-In Data mining there are lots of methods are used to detect the outlier by making the clusters of data and then detect the outlier from them. In general Clustering method plays a very important role in data mining. Clustering means grouping the similar data objects together based on the characteristic they possess. Outlier Detection is an important issue in Data mining; particularly it has been used to identify and eliminate anomalous data objects from given data set where outlier is the data item whose value falls outside the bounds in the sample data may indicate anomalous data. In this work we have suggested a clustering based outlier detection algorithm for effective data mining which uses enhanced k-means clustering algorithm to cluster the data sets and weight based center approach. In proposed approach, two techniques are combined to efficiently find the outlier from the data set. Threshold value can be calculated programmatically by taking absolute value of minimum and maximum value of a particular cluster. The experimental results demonstrate that enhanced method takes least computational time and concentrates on reducing the outlier that could improve efficiency of k-means clustering for achieving the better quality clusters.