The value of multimodal data in classification of social and emotional aspects of tutoring

Cukurova, Mutlu; Kent, Carmel and Luckin, Rosemary (2019). The value of multimodal data in classification of social and emotional aspects of tutoring. In: International Conference on Artificial Intelligence in Education -- AIED 2019: Artificial Intelligence in Education, 25-29 Jun 2019, Chicago, IL, Springer, pp. 46–51.

URL: https://link.springer.com/chapter/10.1007/978-3-03...

Abstract

There are many aspects of tutoring that are associated with social and emotional learning. These are complex processes that involve dynamic combinations of skills, abilities and knowledge. Here, we present the results of our investigation on the particular personal, emotional, and experience traits of tutors who are likely to be successful at social and emotional aspects of tutoring. In particular, we present our approach to measure the social and emotional aspects of tutoring through classification models of 47 candidates' multimodal data from audio and psychometric measures. Moreover, we compare the accuracy of models with unimodal and multimodal data, and show that multimodal data leads to more accurate classifications of the candidates. We argue that when evaluating the social and emotional aspects of tutoring, multimodal data might be more preferrable.

Viewing alternatives

Item Actions

Export

About

  • Item ORO ID
  • 80144
  • Item Type
  • Conference or Workshop Item
  • ISBN
  • 3-030-23203-4, 978-3-030-23203-0
  • Extra Information
  • Also part of the Lecture Notes in Artificial Intelligence book sub series (LNAI, volume 11625)
  • Keywords
  • multimodal data; social and emotional learning; tutoring
  • Academic Unit or School
  • Faculty of Wellbeing, Education and Language Studies (WELS)
  • Copyright Holders
  • © 2019 Springer Nature Switzerland AG
  • Depositing User
  • Carmel Kent

Recommendations