|International Journal of Applied Information Systems
|Foundation of Computer Science (FCS), NY, USA
|Volume 2 - Number 4
|Year of Publication: 2012
|Authors: Meenakshi Pc, Meenu S, Mithra M, Leela Rani P
Meenakshi Pc, Meenu S, Mithra M, Leela Rani P . Fault Prediction using Quad Tree and Expectation Maximization Algorithm. International Journal of Applied Information Systems. 2, 4 ( May 2012), 36-40. DOI=10.5120/ijais12-450338
The objective of the paper is to predict faults that tend to occur while classifying a dataset. There are various clustering algorithms that prevail to partition a dataset by some means of similarity. In this paper, a Quad Tree based Expectation Maximization (EM) algorithm has been applied for predicting faults in the classification of datasets. K-Means is a simple and popular approach that is widely used to cluster/classify data. However, K-Means does not always guarantee best clustering due to varied reasons. The proposed EM algorithm is known to be an appropriate optimization for finding compact clusters. EM guarantees elegant convergence. EM algorithm assigns an object to a cluster according to a weight representing the probability of membership. EM then iteratively rescores the objects and updates the estimates. The error-rate for K-Means algorithm and EM algorithm are computed, denoting the number of correctly and incorrectly classified samples by each algorithm. Result consists of charts showing on a comparative basis the effectiveness of EM algorithm with quad tree for fault prediction over the existing Quad Tree based K-Means (QDK) model.