Prediction of Students Performance for a Multi Class Problem Using Naïve Bayes Classifier
A Daveedu Raju1, Ch Gaayathre2, G Leela Deepthi3, K Madhuri4, D Maheswari5
1A Daveedu Raju, Pursuing B.Tech, Ramachandra College of Engineering, Eluru (Andhra Pradesh), India.
2Ch Gaayathre, Pursuing B.Tech, Ramachandra College of Engineering, Eluru (Andhra Pradesh), India.
3G Leela Deepthi, Pursuing B.Tech, Specialization Computer Science and Engineering Eluru (Andhra Pradesh), India.
4K Madhuri, Pursuing, 3rd B.Tech, Specialization Computer Science and Engineering, at Ramachandra College of Engineering, Eluru (Andhra Pradesh), India.
5D Maheswari, Pursuing, 3rd B.Tech,Specialization Computer Science and Engineering, at Ramachandra College of Engineering, Eluru (Andhra Pradesh), India.
Manuscript received on 01 May 2019 | Revised Manuscript received on 15 May 2019 | Manuscript published on 30 May 2019 | PP: 2209-2214 | Volume-8 Issue-7, May 2019 | Retrieval Number: G5668058719/19©BEIESP
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: Many Engineering Colleges in India are competing with one another to improve the standards of the college by providing best education to the students. The major marking sign among the various criteria is the pass percentage of the students. Sometimes the college management is miffed by stack holders such as parents, other professional bodies, alumni due to the pressure impounded on the students to get the best pass percentage. The proposed paper uses the traditional, but powerful naive Bayes classifier for forecast the student performance, that in turn help the faculty and management to take appropriate movements. The data is collected from the students of 4 year bachelor degree programs of Computer Science and Electronics programs. The data preprocessed for missing value imputation and attribute subset selection. The Bayes classifier model is built by the preprocessed data. The model is tested for check of accuracy and that provided satisfactory results on unknown class label forecasting or prediction, although the features are assumed to be independent as norms of Bayes’ theorem. This helps the teachers and all the stakeholders of the academic institutions that lead to know the performance of the students and to give them the knowledge based on their performance. Further the students and the stakeholders can take corrective actions against the students, whose result is dissatisfactory and it helps to improve their result.
Keyword: Naïve Bayes Classifier, Prediction Preprocessing, Student Performance
Scope of the Article: Regression and Prediction.