Abstract:
Education is a key factor for achieving continuing economic progress. Many students are getting
less in result many reasons. More academic institutes now store massive student educational and
related data. A large number of students enter every year. So, the demanding growth of data in
education sectors continues. Handling and analyzing such a high amount of raw data manually
for performance evaluation creates dissatisfaction, boring and unsuccessful. Manually handling
and evaluating such a big amount of data for performance evaluation leads to discontent,
boredom, and failure. To put it another way, traditional methodologies are overly complex and
difficult to analyze and evaluate. To solve these issues, we use data mining techniques, which are
a set of machine learning algorithms, to examine the data.In other words automated discovery of
previously unknown, valid, novel, useful, and understandable patterns in school databases.The
study's major goal is to use data mining to create a predictive model for student academic
performance in Amhara National Regional State's Awi zone Injibara and AgewMidir general
secondary schools. This can greatly support policymakers, planners, and education providers
working on the control of student performance. The methodology used for this research was a
hybrid six-step CRISP Knowledge Discovery Process.The essential data were gathered from a
school data warehouse created specifically for student result purposes, collected from 2009 to
2012 E.C. The researcher used two popular data mining algorithms (J48 Decision Trees and
Naïve Bayes Classifier) to develop the predictive model using a larger dataset (5408 cases). The
researcher used 10-fold cross-validation and percentage split test mode for data mining
algorithms of the two predictive models for performance comparison purposes. The results
indicated that the decision tree (J48 algorithm) is the better predictor with a prune d parameter
of the tree of 10-fold cross-validation mode due to the nature of data which is categorical for
which J48 is better. It has 95.37% accuracy on the given school dataset with j48 algorithm,
Naïve Bayes Classifier came out to be the second with an accuracy of 91.76%.
Keywords: Education, Student performance, Data mining, Classification, Weka, Decision Tree,
Naive Bayesian