| Autor: |
Blobstein, Ariel, Gal, Kobi, Karger, David, Facciotti, Marc, Hyunsoo Kim, Almahmoud, Jumana, Kamali Sripathi |
| Přispěvatelé: |
Mitrovic, Antonija, Bosch, Nigel |
| Rok vydání: |
2022 |
| Předmět: |
|
| DOI: |
10.5281/zenodo.6853101 |
| Popis: |
Emoji are commonly used in social media to convey attitudes and emotions but have rarely been studied in educational contexts. This paper provides a data-driven study of students' use of emoji in an annotation-based Biology course forum where students' comments are anchored in the course material. The 11 emoji in the forum were hand-designed by the instructors to allow students to communicate diverse types of affects related to the course material, such as inviting discussion about a topic, declaring a topic is interesting, or requesting assistance about a topic. We analyze emoji usage in the course by hundreds of students, showing that some emoji frequently appear together in posts relating to the same paragraphs, communicating complex affective states by students. We explore the use of computational models for predicting emoji at the post level, which can allow teachers to infer information about students' affective states directly from students' posts, even when they are lacking emoji. We show that partitioning the emoji into distinct groups has pedagogical benefits to instructors while also improving performance when predicting emoji-groups rather than individual emoji. Our approach can be generalized to other courses and potentially improve teachers' ability to make sense of students' affect. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
