Educational Data Classification and prediction using Data Mining Algorithms
J. Jayapradha1, Kishore Jagan Jothi Kumar2, Binti Deka3
1Jayapradha J, Assistant Professor of SRMIST, Chennai.
2Kishore Jagan Jothi Kumar, Master of Computer Science at Arizona State University, Tempe, AZ, USA .
3Binti Deka, Case Manager in Onco.com, Banglore, KA, India.
Manuscript received on 13 August 2019. | Revised Manuscript received on 18 August 2019. | Manuscript published on 30 September 2019. | PP: 8674-8678 | Volume-8 Issue-3 September 2019 | Retrieval Number: C6457098319/2019©BEIESP | DOI: 10.35940/ijrte.C6457.098319
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© 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: Data Mining is the process of extraction interesting patterns from huge data sets and converts the patterns into logical structure for further Analysis. Predictive Modeling processes that make use of data mining, Machine learning and probability methods to forecast. Engineering is the most widely accepted stream of education in India. Students are uncertain about which department to join in engineering. It is important to improve the individual performance and help the students make the perfect choice regarding the department. In this paper, the hidden information from the previously recorded enrollment details during admission process is used to solve the students’ uncertainty in their choice of department. In addition to this, the performance of alumnae also needs to be analyzed by the teachers to have a clear idea about the future of existing students. Our main goal is to unravel these problems using predictive Modeling. Here, we are focusing on three classification algorithms namely, support vector machine, Random Forest and Naïve Bayes. Data has been collected, normalized and applied to the three different classification algorithms, from which the best model is formulated using various parameters of evaluation. In this paper, we present our approach towards implementing the best model which is built based on the profession of parents, demographic features, type of location of the student and correlation between high school and higher secondary examinations. The Result of this research work shows that Random forest is efficient for the data set used when compared to the other two Classification algorithms.
Keywords: Predictive Modeling, Classification Algorithm, Support Vector Machine, Random Forest, Naïve Bayes, Data Mining.
Scope of the Article: Classification