A Method for Converting Lecture Videos into Microcontents and Visualizing Them

With the spread of online courses, a large amount of lecture videos has become available. In order to make more advanced use of existing lecture videos, it is necessary to convert the videos into microcontents and visualize them in a way that allows users to easily find the parts they need. For this purpose, in this paper, we propose a method for converting lecture videos into microcontents. The method vectorizes transcripts of videos using Doc2Vec and then divides videos based on the distance between these vectors. In addition, we compared visualization methods using principal component analysis, multidimensional scaling, and t-SNE, and found that t-SNE is suitable for the visualization.