A New Way of Visualizing Curricula Using Competencies: Cosine Similarity and t-SNE
Kunihiko Takamatsu, Katsuhiko Murakami, Yasuhiro Kozaki, Kenya Bannaka, Ikuhiro Noda, Raphael-Joel Wei Lim, Kenichiro Mitsunari, Tadashi Nakamura, Yasuo Nakata · 2018
This paper outlines a new way of visualizing curricula using competencies via a combination of cosine similarity, t-distributed stochastic neighbor embedding (t - SNE) methods, and scatter plotting. We have already published a report on the use of multidimensional scaling with the same methods. In this report we show that a t-distributed stochastic neighbor embedding method is more useful than a multidimensional scaling method. We believe that this visualization will be useful when students select their courses.