ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

FINAL SEMESTER MARKS PREDICTION USING CLASSIFICATION TECHNIQUES

Journal: International Journal of Advanced Research in Engineering and Technology (IJARET) (Vol.11, No. 08)

Publication Date:

Authors : ;

Page : 32-43

Keywords : Data Mining (DM); Educational Data Mining (EDM); WEKA; Prediction; Classification.;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

Research in the field of academics is developing expeditiously in various areas one of the areas is of Educational Data Mining. During the COVID-19 pandemic, most of the educational institutions are closed all around the world. Examinations are not being conducted. There are not so many applications or modules available for online examinations. So, it becomes difficult for the students to appear in the final semester examinations, and hence the declaration of results and awarding corresponding Diplomas and Degrees to the Students. Taking into the consideration the present state and apprehension of various students studying in various Government and Private Polytechnic colleges (Diploma Level) of Punjab research study has been conducted to answer two research questions arising, first one how to predict performance of students in terms of marks in the final semester exam? And second how to foretell or project the marks obtained in the final semester by each student toward the end of the semester? To answer the above questions, different Classification algorithms namely; k-NN, Linear Regression, Multilayer Preceptor, and Decision Tree were applied after data pre-processing, and filtering the most influential features and removing the irrelevant and hence reducing dimensionality and hence reducing the overall complexity of the model. After training and testing the most accurate model was found to be 1BK. And we were able to find the answer to both the research questions and able to make the predictions of the results of the 6 thsemester of the students. As a future direction effort will be made to further reduce the error and include the other features related to student personal, demographic, and socio-economic factors.

Last modified: 2021-02-20 13:53:22