Authors :
Suleiman Sahabi; Anas Shehu; Shamsu Sani; Abubakar Sani
Volume/Issue :
Volume 8 - 2023, Issue 3 - March
Google Scholar :
https://bit.ly/3TmGbDi
Scribd :
https://bit.ly/40jjnqs
DOI :
https://doi.org/10.5281/zenodo.7749680
Abstract :
This research was conducted in data mining. To
our knowledge, no research covered seven programs of
higher institutions for data mining purposes. This work used
real dataset of students from seven department of Kebbi
State Polytechnic Dakin-gari. Classification, association and
clustering were used to discover hidden patterns in the
dataset. WEKA workbench was used to run the experiment
and evaluate the results. Classification was done with
optimum accuracy where four classes were identified i.e.
weaker, weak, good and better students. After association
rule was done, the authors found out strong correlation
between (ATT and CA) with (GPA), (EF and CA) with
(GPA) and (EF and CA) with (CO). And that affect the
students in their GPA results. Same test data was used for
hierarchical clustering where two clusters were returned and
162 tuples was distributed between the clusters in 81% and
19% fashion, Conclusively, the authors strongly feel that the
management of Kebbi State Polytechnic Dakin-gari with a
matter of urgency need to tackle these discovered pattern to
minimize rate of failure and drop out within the students.
Keywords :
WEKA, Data Mining Tool, Dakin-Gari, Clustering, Association, Classification, Metric.
This research was conducted in data mining. To
our knowledge, no research covered seven programs of
higher institutions for data mining purposes. This work used
real dataset of students from seven department of Kebbi
State Polytechnic Dakin-gari. Classification, association and
clustering were used to discover hidden patterns in the
dataset. WEKA workbench was used to run the experiment
and evaluate the results. Classification was done with
optimum accuracy where four classes were identified i.e.
weaker, weak, good and better students. After association
rule was done, the authors found out strong correlation
between (ATT and CA) with (GPA), (EF and CA) with
(GPA) and (EF and CA) with (CO). And that affect the
students in their GPA results. Same test data was used for
hierarchical clustering where two clusters were returned and
162 tuples was distributed between the clusters in 81% and
19% fashion, Conclusively, the authors strongly feel that the
management of Kebbi State Polytechnic Dakin-gari with a
matter of urgency need to tackle these discovered pattern to
minimize rate of failure and drop out within the students.
Keywords :
WEKA, Data Mining Tool, Dakin-Gari, Clustering, Association, Classification, Metric.
