Summarization of Learning Materials Using Digital Notes and Online Data
| Autor: | Chen-YuLin, 林真瑜 |
|---|---|
| Rok vydání: | 2018 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 106 Most students take notes when they are in class, but in most cases, they couldn’t write down the whole contents during class. Students can get the missing parts if they share their notes with each other. This study refers to the idea of the sharing economy and links the students' notes to complete the missing parts of each one. Students and teachers are getting used to electronic teaching materials. The emphases of courses are often integrated into slides. It’s difficult to understand what teachers want to convey just by notes or slides. However, the contents of the textbook are so tedious that students cannot get the point easily. Furthermore, what teachers emphasize is sometimes just mentioned without going into deeply in the textbook, so that students need to search for relevant information, which takes much time to filter out unhelpful information. Therefore, this paper proposed a method called NoteSum to integrate and summarize learning materials automatically, which will help students learn. This study conducted several experiments and got the following conclusions. First, the Jensen-Shannon (JS) Divergence was used to assess the summary, and those generated for the discussion course performed better. Next, the three attributes—the presence of topic terms, the number of non-topic words, and the ratio of the words with important part of speech—had different effects on different subjects. At last, we compared NoteSum with other summarization systems through automatic and manual evaluation and verified that NoteSum was more powerful and could help students learn more quickly. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|