TY - GEN
T1 - Effect of Presenting Co-occurrence Networks that Reflect the Activeness of Face-to-Face Discussions
AU - Muraoka, Taisei
AU - Ishikawa, Naruaki
AU - Ozawa, Shigeto
AU - Egi, Hironori
N1 - Publisher Copyright:
© 2021, Springer Nature Switzerland AG.
PY - 2021
Y1 - 2021
N2 - This paper proposes a system that presents outlines of discussions in other groups in the same classroom as co-occurrence networks that reflect the activeness of discussions. The proposed system focuses on discussions conducted by two people during a lecture. The activeness of the discussions is analyzed using nonlinguistic acoustic information, which is calculated based on participants utterances collected by wearable devices. The co-occurrence networks are drawn with emphasizing the active parts of discussions by each group. We conducted an experiment to verify the effect to the participants by applying the proposed system to a lecture, and the effect of presenting the co-occurrence networks was examined in a subjective evaluation. The participants conducted another discussion after observing the co-occurrence networks of the previous discussion, and the second discussion was scored by the lecturer. As a result, the co-occurrence networks reflecting discussion activeness were not evaluated higher than those that did not reflect discussion activeness. This suggests that the variety of topics in the co-occurrence networks may stimulate discussion participants more effectively.
AB - This paper proposes a system that presents outlines of discussions in other groups in the same classroom as co-occurrence networks that reflect the activeness of discussions. The proposed system focuses on discussions conducted by two people during a lecture. The activeness of the discussions is analyzed using nonlinguistic acoustic information, which is calculated based on participants utterances collected by wearable devices. The co-occurrence networks are drawn with emphasizing the active parts of discussions by each group. We conducted an experiment to verify the effect to the participants by applying the proposed system to a lecture, and the effect of presenting the co-occurrence networks was examined in a subjective evaluation. The participants conducted another discussion after observing the co-occurrence networks of the previous discussion, and the second discussion was scored by the lecturer. As a result, the co-occurrence networks reflecting discussion activeness were not evaluated higher than those that did not reflect discussion activeness. This suggests that the variety of topics in the co-occurrence networks may stimulate discussion participants more effectively.
KW - Co-occurrence network
KW - Group discussion
KW - Nonlinguistic acoustic information
KW - Word embedding
UR - http://www.scopus.com/inward/record.url?scp=85112148312&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85112148312&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-77889-7_24
DO - 10.1007/978-3-030-77889-7_24
M3 - Conference contribution
AN - SCOPUS:85112148312
SN - 9783030778880
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 347
EP - 360
BT - Learning and Collaboration Technologies
A2 - Zaphiris, Panayiotis
A2 - Ioannou, Andri
PB - Springer Science and Business Media Deutschland GmbH
T2 - 8th International Conference on Learning and Collaboration Technologies, LCT 2021, held as Part of the 23rd International Conference, HCI International 2021
Y2 - 24 July 2021 through 29 July 2021
ER -