I. (10) that described the design and implementation of

                                                                                                                                                I. review of literature

Recently several studies have
incorporated data mining into academic research. A study (7) used clustering
and decision rule algorithm to get knowledge about student’s academic
achievement and students retention and concluded that more variables were
required for a more realistic analysis of student’s performance.  Decision rule was also used in a study (8)
that built a model to justify the capability of data mining in higher
education. The authors were able to gain a better understanding of student
enrollment patterns and were able to describe a roadmap for the application of
datamining in predicting which students were less likely to be successful in
the course.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

This study (9) used Bayesian
networks to predict graduating cumulative Grade Point Average based on
applicant background at the time of admission.  Attributes from the admission data like age,
gender, marital status, nationality, English test score, institute of the
previous degree, major of the previous degree, GPA of the previous degree, field
of study and degree program to which the applicant is applying were processed
by clustering the values into groups of similar majors and similar fields of
study. Attributes from the socio-economic environment data like can Gross
National Income (GNI), LIC (Low Income), LMC (Low Middle Income), UMC (Upper
Middle Income), NOC (High Income, non-OECD), and OEC (High Income, OECD) were
also considered. The study concluded that socio-economic environment can play a
major role in the performance of students.

Association rules and Apriori
algorithm was used in a study (10) that described the design and implementation
of a decisional tool dedicated to an institution that hopes to improve the
quality of its service.  Authors analyzed
the factors that determined the success and failures of students and attempt to
increase success chances of students. The result of this study was an
implemented tool that could, for example, discover the tendency of a teacher to
give bad or good marks and also reveal the difficulties met by the students in
certain modules with regard to the others. Apriori algorithm was an improved
algorithm for a study that mined a student’s data table (11) which implies that
in practices teachers can predict scores of other courses from scores of some
courses and speak or act with a well-defined objective in mind.

A combination of clustering,
association rules and decision trees were used in a study (12) that aimed to
discover patterns within historical students’ academic and financial data. It
analyzed students’ academic environment, dropout rate and financial behavior. Results
showed a concentration on three main attributes; students’ academic
achievements, students’ drop out, and students’ financial behavior.

Ayesha et al (13) used K-Means
clustering to gain information that comes from student’s evaluation data like
class quizzes mid and final exam assignment to determine the performance of
students and predict weak students. The results from this study will help
teachers reduce the drop out ratio to a significant level.