الفهرس | Only 14 pages are availabe for public view |
Abstract The process of identifying the subjects that are taught in a particular scientific course is considered a difficult process that is carried out by the academic course coordinator in a manual manner, in which he depends on his experience as well as the Intended Learning Outcomes (ILO{u2019}s) of such course. The aim of this thesis is to provide an approach for automatic selection of topics for various courses. The proposed approach is based on N-gram analysis through which relevant and critical phrases are extracted from targeted references with their corresponding word frequencies. Then, proper candidate topics are selected according to the ILO{u2019}s of the considered course. The proposed approach has been evaluated by automatically generating two different courses based on the same specific material based on their different ILO{u2019}s. It is found that the proposed approach can be used also for fixing some currently existing graduate or postgraduate courses specifications. The results indicated that, the first course (Cryptography and Network Security) the automatic topics selection approach achieves 94% of the topics that defined manually in the course specifications topics. In addition, it recommends one extra chapter to be included in the course specifications topics, and the second course (Advanced Topics in Database Design & Implementation) the automatic topics selection approach achieves 90% of the topics that defined manually in the course specifications topics |