Exam-based education system


Xuhang Song

Date of Award


Degree Name

M.S. in Computer Science


Department of Computer Science


Advisor: James P. Buckley


Big data is growing in importance in everyday life, yet traditional models of University education do not make good use of it. This thesis proposes a system that allows students to find courses based on similarity measures and take these courses in an exam-based environment. We describe a new mining method that can efficiently search, cluster and perform related functions in the system. The basic idea of this mining is to map courses in a university to buildings in a city. This means that finishing a degree or getting a skill is analogous to finding a path in the city. A number of approaches to build the city are presented. In the process of developing an algorithm, we use machine learning, artificial intelligence, and classic mining methods as core ideas.


Universities and colleges Curricula, Self-managed learning, Data mining, Big data, Computer Science, Data mining, Genetic Agortihem, Education

Rights Statement

Copyright 2014, author