Literaturnachweis - Detailanzeige
Autor/inn/en | Zhu, Jile; Li, Xiang; Wang, Zhuo; Zhang, Ming |
---|---|
Titel | An Effective Framework for Automatically Generating and Ranking Topics in MOOC Videos [Konferenzbericht] Paper presented at the International Conference on Educational Data Mining (EDM) (10th, Wuhan, China, Jun 25-28, 2017). |
Quelle | (2017), (6 Seiten)
PDF als Volltext |
Sprache | englisch |
Dokumenttyp | gedruckt; online; Monographie |
Schlagwörter | Large Group Instruction; Online Courses; Educational Technology; Technology Uses in Education; Video Technology; Data Collection; Data Analysis; Lecture Method; Scaling; Feedback (Response) |
Abstract | Although millions of students have access to varieties of learning resources on Massive Open Online Courses (MOOCs), they are usually limited to receiving rapid feedback. Providing guidance for students, which enhances the interaction with students, is a promising way to improve learning experience. In this paper, we consider to show students the emphasis of lectures before their learning. We propose a novel framework that automatically generates and ranks the topics within the upcoming chapter. We apply the Latent Dirichlet Allocation (LDA) model on the subtitles of lectures to generate topics. We then rank the importance of these topics through a particular PageRank method, which also leverages structural information of lectures. Experimental results demonstrate the effectiveness of our approach, with a 18.9% improvement in Mean Average Precision (MAP). At last, we simulate two cases to discuss how can our framework guide students according to their learning status. [For the full proceedings, see ED596512.] (ERIC). |
Anmerkungen | International Educational Data Mining Society. e-mail: admin@educationaldatamining.org; Web site: http://www.educationaldatamining.org |
Erfasst von | ERIC (Education Resources Information Center), Washington, DC |
Update | 2020/1/01 |