The skills required of new employees by industry are increasingly interdisciplinary and creativity-related because of a paradigm shift in target markets. Engineering education should therefore focus on helping students develop their creativity and critical thinking skills. A student's level of creativity is usually evaluated by examining his or her final projects. However, the language that students use in discussions and interactions can be analyzed to determine their cognitive processes and thus their creativity. This study collected 1 year of records of discussions and interactions on a Moodle learning platform among students in two college courses (Computer Science and Engineering). The discussions and interactions were filmed and recorded in a backend database and were later transcribed. The transcripts were arranged and analyzed. The data were divided into a training set (79 discussions; 90%) and a test set (9 discussions; 10%) before data mining was performed. The training set was used to construct a training model, and the test set was employed to examine whether the proposed model correctly predicted creativity in the students. K-means clustering was used to cluster the language in the discussion content. The level of creativity of each student was correctly predicted by the model, which can be used by teachers to provide feedback and support in a timely manner for triggering different thinking in students to enhance his or her creative thinking. The proposed model can thus identify level of creativity and assist both teachers and students.
|期刊||ASEE Annual Conference and Exposition, Conference Proceedings|
|出版狀態||Published - 2018 六月 23|
|事件||125th ASEE Annual Conference and Exposition - Salt Lake City, United States|
持續時間: 2018 六月 23 → 2018 十二月 27
All Science Journal Classification (ASJC) codes
- 工程 (全部)