Takayuki Sekiya, Yoshitatsu Matsuda, Kazunori Yamaguchi
Proceedings - Frontiers in Education Conference, FIE 2017- 1-9 2017年12月12日 査読有り
For university students, a syllabus gives fundamental information about a course, and is important for choosing a course. However, it is not an easy task for students to grasp the topics actually covered by a course syllabus because they have only little knowledge about topic words in the syllabus before they learn the course. We have been studying on a machine learning method of systematically analyzing syllabi by standard curricular guidelines such as "Computing Science Curricula CS2013," which is released by the ACM and IEEE Computer Society. We acquired a probabilistic topic model of computer science syllabi, and developed a tool for investigating the actual syllabi in the model. In this paper, we introduce a web-based tool and demonstrate its effectiveness by some examples. By applying our tool to a syllabus, students and teachers can know how strongly the syllabus and topics are related quantitatively, where each topic corresponds to the Knowledge Area of CS2013 such as "Algorithms and Complexity (AL)." In addition, the tool utilizes four meta-topics (HUMAN, THEORY, PROGRAMMING, and SYSTEM), which are extracted by investigating the actual syllabi. The tool also provides a list of syllabi similar to the given syllabus, which are selected from the actual syllabi of the top-ranked universities. These information are beneficial for students to understand the courses and for teachers to improve their syllabi.