Outlier Detection using Clustering Techniques – K-means and K-median

Outlier detection in data mining helps to identify and dispose of irregular data objects from the given dataset. In this work, various clustering and outlier techniques are initially reviewed. Secondly, the problem is identified using the K-means clustering algorithm for outlier detection Finally, an optimal solution is proposed from K-median clustering and compared it with the weight-based K-mean grouping. To test the algorithm of outlier detection, the real dataset is considered and the above process is simulated in MATLAB. The experimental result is compared with the proposed K-median weight-based approach and has the maximum weight value in the frame.