Abstract
Associative classification (AC) is a branch in data mining that utilises association rule discovery methods in classification problems.This paper idea aims to discuss and evaluate a modeling approach for student evolution. It is developed as a component of an adaptive achievement system. At the beginning of the process, associations of student achievement results are found based on each student’s factors that affect learning process, which finds the relationship between student evolution during years of study and understanding the modular scheme, that finds the main effect of enrolling on the correct modules i.e. getting the right advice and student support regarding choosing modules, meeting all the necessary prerequisites, having summer courses, and taking in consideration the student’s high school grades, as well as finding the relationship between modules type and student gender . Clustering [10], or unsupervised classification, method is employed to model this task. The goal of clustering is [7] to objectively partition data into homogeneous groups such that the within group object similarity and the between group object dissimilarity are determined. Clustering here is used to model student achievement according to predefined criterion functions that measure similarity among students who grant certain goal having the same conditions using data collected from University Database. A clustering method is developed for this step. We evaluated the student progress according to associations between different factors using data collected. We concluded the performance of those groups using these two approaches. Now, the need for solid information about student evolution and how to improve it has only grown in importance for state policy. The compelling metaphor of increasing flow through the “educational pipeline” is now common in state policy discussions, fueled by more vocal recognition by business and civic leaders of the importance of the critical “supply chain” of educational capital in their states.