© 2014, IJARCSMS All Rights Reserved 117 | P a g e ISSN: 232 7782 (Online) 1 Computer Science and Management Studies  International Journal of Advance Research in Volume 2, Issue 9, September 2014 Research Article / Survey Paper / Case Study Available online at: www.ijarcsms.com An Experimental analysis of Parent Teacher Scale Involvement with help of K Mean Clustering Technique using Matlab Dr. Manju Kaushik 1 Associate Professor JECRC University Jaipur – India Bhawana Mathur 2 Research Scholar JECRC University Jaipur – India Abstract: Researcher collect data through Parent Teacher Scale Involvement psychology Education Test in the form of Questionnaire. Researcher collect random sample of 28 students’ data from School, Jaipur as a random selection of Students of Class 1stto 10 th Class. To make cluster, Find Distance point from the cluster, Find Minimum distance from each point of data sets. Assign Points Pi to Centroids Ck. Update centroids .Do Iteration until centroid converges or uses loop. Researcher makes Questionnaire of 34 questions. Three parameters are like these School involvement, home involvement, parent teacher school involvement. With the help of K Means Technique, using Matlab Object Oriented Software System. Through K Means Methodology Computation Large data sets various parameters, Questionnaire’s Question, and their values after computational K Means ,Optimized the results in object oriented system like Matlab. Keywords: Questionnaire cluster, Distance point, Minimum distance Centroids, data sets, School Involvement. I. INTRODUCTION The primary objective of the proposed clustering methodology is to provide a general but illuminating view of a software system that may lead engineers to useful conclusions concerning its maintainability. This data mining technique is useful for Similarity/Dissimilarity analysis; in other words it analyzes what data points are close to each other in a given dataset. This way, mutually exclusive groups of classes are created, according to their similarities and hence the system comprehension and evaluation is facilitated. Thus, maintenance engineers are provided a panoramic view of a system's evolution, which helps them in revising the system's maintainability, studying the classes' behavior from version to version and discovering programming patterns and "unusual" or outlier cases which may require further attention. An approach is to combine all the data sets (the data points corresponding to classes) into a large data set. II. LITERATURE REVIEW “Improving the Accuracy and Efficiency of the k-means Clustering Algorithm”, an improvement in K-means clustering is shown. The first phase of K-means clustering algorithm, the initial centroids are determined systematically so as to produce clusters with better accuracy 1. “An Efficient K-Means Clustering Algorithm for Reducing Time Complexity using Uniform Distribution Data Points”, the uniform distribution of the data points is discussed that how this approach reduce the time complexity of the K-means clustering algorithm 2. By using this approach the elapsed time is reduced and the cluster is of better quality. A very good method is used for finding the initial centroid. In this initially, the distance between each data points is computed. “An Iterative Improved k-means Clustering” discuss an iterative approach which is beneficial in reducing the number of iterations from k-mean algorithm, so as to improve the execution time or by reducing the total number of distance calculations 3. So Iterative improved K-means clustering produces good starting point for the K-means algorithm instead of selecting them randomly. And it will lead to a better cluster at the last result.