Loading

An Effectual Ga Based Association Rule Generation and Fuzzy Svm Classification Algorithm for Predicting Students Performance
E.Chandra Blessie1, K R Vineetha2

1E.Chandra Blessie, Professor, Department Of Computer Application, Nehru College of Management, Coimbatore.
2K R Vineetha, PhD Research Scholar, Nehru College of Management, Coimbatore.
Manuscript received on July 20, 2019. | Revised Manuscript received on August 10, 2019. | Manuscript published on August 30, 2019. | PP: 2915-2920 | Volume-8 Issue-6, August 2019. | Retrieval Number: F8805088619/2019©BEIESP | DOI: 10.35940/ijeat.F8805.088619
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: This investigation provides outcome of utilizing educational data mining [EDM] to design academic performance of students from real time and online dataset collected from colleges. Data mining is determined to examine non-academic and academic data; this model utilizes a classification approach termed as Fuzzy SVM classification with Genetic algorithm to attain effectual understanding of association rule in enrolment and to evaluate data quality for classification, which is identified as prediction task of performance and academic status based on low academic performance. This model attempts to predict student’s performance in grading system. Academic and student records attained from process were considered to train models estimated using cross-validation and formerly records from complete academic performance. Simulation was performed in MATLAB environment and show that academic status prediction is enhanced while hybrid dataset are added. The accuracy was compared with the existing models and shows better trade off than those methods.
Keywords: Educational Data Mining, Fuzzy SVM, Genetic, academic performance, academic records