Knowledge based student academic advising in institutions of higher learning in kenya

Abstract

Technological advances in the last two decades have led to reduced costs of computer hardware and software. This has in turn led to widespread acquisition and use Student Management Information Systems by institutions of higher learning resulting in huge databases of student data. Despite these advancements, reporting applications have remained largely rudimentary employing simple reports such as a student transcript, a consolidated mark sheet and so on.

Data mining techniques can be used to obtain knowledge from large collections of data by employing techniques from both computer science and statistics thereby enriching the reporting applications. These techniques can be used to provide patterns and analyses that cannot be obtained using rudimentary querying and reporting techniques. Furthermore these patterns can be used to build models that can be used to perform predictions among many other applications.

This research proposes to employ data mining techniques to build predictive models that can be used to predict the degree honors class that a student is likely to get. Using this knowledge plus that obtained using clustering and classification data mining techniques, prudent advice can be furnished to students early enough about their likely final degree honors class. With this knowledge, students can then be advised on how to improve their performance using the knowledge provided by classification and clustering algorithms.