Multi-Modal Affective Computing Technology Design the Interaction between Computers and Human of Intelligent Tutoring Systems

In this paper, the authors are using emotion recognition in two ways: facial expression recognition and emotion recognition from text. Through this dual-mode operation, not only can strength the effects of recognition, but also increase the types of emotion recognition to handle the learning situati...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:International journal of online pedagogy and course design Ročník 6; číslo 1; s. 13 - 28
Hlavní autoři: Lin, Hao-Chiang Koong, Su, Sheng-Hsiung, Wang, Cheng-Hung, Huang, Zu-Ching
Médium: Journal Article
Jazyk:angličtina
Vydáno: IGI Global 01.01.2016
Témata:
ISSN:2155-6873, 2155-6881
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In this paper, the authors are using emotion recognition in two ways: facial expression recognition and emotion recognition from text. Through this dual-mode operation, not only can strength the effects of recognition, but also increase the types of emotion recognition to handle the learning situation smoothly. Through the training of image processing to identify facial expression, the emotion from text is identifying by emotional keywords, syntax, semantics and calculus with logic. The system identify learns' emotions and learning situations by analyzing, choosing the appropriate instructional strategies and curriculum content, and through agents to communicate between user and system, so that learners can get a well learning. This study uses triangular system evaluation methods, observation, questionnaires and interviews. Experimental design to the subjects by the level of awareness on art and non-art to explore the traditional teaching, affective tutoring system and no emotional factors learning course website these three kinds of ways to get results, analysis and evaluate the data.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:2155-6873
2155-6881
DOI:10.4018/IJOPCD.2016010102