ANALYZING STUDENT PERFORMANCE USING ROBUST CLUSTERING TECHNIQUE

Data mining is the method of determining patterns in large datasets with artificial intelligence, machine learning, statistics and database systems. The main goal of higher education institution is to employ data mining methodologies for learning student's performance in the educations. Data mining of ers many tasks that are used to analysis the student performance. The classification task is designed applied to calculate student's performance. In addition to, many approaches are used for data classification to support decision tree method. The decision tree technique is employed to accurately predict the student performance. The existing work presented a SVM Prediction technique for evaluating the student Grade Point Average (GPA) in computer education and instructional technology at the end of first, second, and third-year courses. Three kinds of procedures are involved in SVM prediction data mining technique that is data preparation, formulation of prediction model and evaluation of the SVM prediction model. By using linear arithmetic, the SVM prediction model performs tasks of classification and regression using linear combination of features based on variables. However, the SVM technique does not provide suggestive methods for enhancing the student GPA. This technique fails to determine the exact recollect values and does not carried out subjective analysis. To overcome these drawbacks, the proposed work presents Spectral Cluster based Decision Tree Data Mining Technique for analyzing Student Performance in Higher Education Institutions. To develop Cluster based Decision Tree technique predicts exact student GPA with aiming at improves true positive values. In order to provide decision tree based suggestive methods which helps to increase the GPA of weak students. The performance measure of proposed technique are done with following metrics such as, Prediction Accuracy, True Positive, False Positive and Number of decision rules. This is also the case when applying Cluster Analysis methods, where those troubles could lead to unsatisfactory clustering results. Robust Clustering methods are aimed at avoiding these unsatisfactory results.