Study Of Fault Prediction Using Quad Tree Based K-Means Algorithm And Quad Tree Based EM Algorithm

Abstract

The paper intends to do a comparative study of the two clustering algorithms, namely K-Means and EM. Quad tree is used as a common algorithm to initialize both the clustering algorithms. The dataset is then clustered and classified separately by K-Means and EM algorithms. The motive of this paper is to prove the effectiveness of EM over K-Means. Classification and clustering of the dataset done via EM is seen to have lower faults as compared to clustering and classification done via K-Means algorithm