Authors :
V. Bakyalakshmi; Dr. S. Kanchana
Volume/Issue :
Volume 8 - 2023, Issue 4 - April
Google Scholar :
https://bit.ly/3TmGbDi
Scribd :
https://bit.ly/42cjtkF
DOI :
https://doi.org/10.5281/zenodo.10250420
Abstract :
Education is a broad and essential subject
that cannot be effectively discussed in a little amount of
space. The amount of data maintained in educational
databases has grown considerably in recent years. The
database contains confidential information regarding
the academic and behavioral progression of students. In
educational settings, the potential to identify student
performance in learning is much important.
Educational Data Mining (EDM) is considered as the
developed research area along with the computational
and psychological methods for predicting the students’
achievements. EDM entails in evaluating the student
data in order to identify hidden student knowledge.
Unbalanced datasets are one of the most critical issues
influencing the performance of classifiers. It is a
significant challenge in the EDM domain that
contributes to inaccurate outcomes. Recently, machine
learning (ML) and deep learning (DL) approaches are
generally developed for tracking and predicting student
achievements by considering different aspects like
student’s academic achievement data, personal data ,
behavior data etc., Predicting student outcomes with
ML and DL methods emphasizes on inferring
information from student achievements data which
helps to comprehend the students' affinity on learning,
adjusting to new challenges or subjects, and
accomplishing the challenges or activities appropriately.
This paper provides a comprehensive review on various
ML and DL frameworks designed to track and predict
student achievements using an online educational
dataset with various student observation variables.
Initially, various frameworks for predicting student
performance based on ML and DL algorithms that have
been developed by numerous researchers are examined
in detail. Then, a comparative analysis is performed to
comprehend the shortcomings of these frameworks and
to provide a new solution to effectively predict the
students' achievements
Keywords :
Educational Data Mining, Student Achievements Machine Learning, Deep Learning, Online Courses
Education is a broad and essential subject
that cannot be effectively discussed in a little amount of
space. The amount of data maintained in educational
databases has grown considerably in recent years. The
database contains confidential information regarding
the academic and behavioral progression of students. In
educational settings, the potential to identify student
performance in learning is much important.
Educational Data Mining (EDM) is considered as the
developed research area along with the computational
and psychological methods for predicting the students’
achievements. EDM entails in evaluating the student
data in order to identify hidden student knowledge.
Unbalanced datasets are one of the most critical issues
influencing the performance of classifiers. It is a
significant challenge in the EDM domain that
contributes to inaccurate outcomes. Recently, machine
learning (ML) and deep learning (DL) approaches are
generally developed for tracking and predicting student
achievements by considering different aspects like
student’s academic achievement data, personal data ,
behavior data etc., Predicting student outcomes with
ML and DL methods emphasizes on inferring
information from student achievements data which
helps to comprehend the students' affinity on learning,
adjusting to new challenges or subjects, and
accomplishing the challenges or activities appropriately.
This paper provides a comprehensive review on various
ML and DL frameworks designed to track and predict
student achievements using an online educational
dataset with various student observation variables.
Initially, various frameworks for predicting student
performance based on ML and DL algorithms that have
been developed by numerous researchers are examined
in detail. Then, a comparative analysis is performed to
comprehend the shortcomings of these frameworks and
to provide a new solution to effectively predict the
students' achievements
Keywords :
Educational Data Mining, Student Achievements Machine Learning, Deep Learning, Online Courses
