Abstract
The number of applicants considered during the admission selection process for universities
has increased exponentially. This has led to development and improvement of the admission
criteria so as to ensure that new intakes that possess the potential to achieve academic success
are selected. The aim of the present study is to examine the relationship between academic
success of architecture student and prior academic performance using K-nearest neighbour
algorithm (k-NN). Data on prior academic performance, which is considered during admission
process, and academic success was collected on four cohorts of architecture students. Then the
data is divided into two parts: training set (70%) and test set (30%). Finally, the k-NN was
developed using the training sets and the predictive performance was evaluated using the test
set. The experimental results shows that the overall accuracy of the k-NN model is 73.33%. It
is anticipated that the developed model could provide useful information that can be used to
identify new intakes whom possess adequate intellectual capabilities to succeed in
undergraduate architecture programs in Nigeria.