A new system for analyzing contents of Computer Science courses
Peerapon Kamlangpuech, Komate Amphawan · 2020
With the growth of technology and the changing of human behavior, the fields of Computer Science (CS), Information Technology (IT), and Data Science (DS) become more popular and should be learned by children, students, and people. Consequently, institutes widely initiate new online/offline courses to serve this need. However, this leads to educate different knowledge and skills in which some of them may not match the needs of companies for hiring employees. To address this issue, we introduce a new system for analyzing content on CS courses, called CSCDA system (Computer Science Course Description Analysis system). The system can identify similar and dissimilar contents belonging to two CS courses by applying text processing techniques and keyword similarity matching. These identified contents can help to set up a standard, to improve integrity and quality, and to reduce redundancy of the contents will be taught in the courses. Experimental studies are conducted on CS courses of Thai Universities to investigate the effectiveness of the CSCDA system based on four measures, i.e. percentage of similar contents, precision, recall, and F-measure, respectively. Last, a comparative study is performed and the result shows that our proposed method can effectively analyze course contents and outperforms the other keyword extraction methods.