Uaffect: Integrating Context Histories to Predict Affective States in Educational Environments

• Sandro O. Dorneles Federal Institute of Rio Grande do Sul http://orcid.org/0000-0001-7821-2351
• Débora Nice Ferrari Barbosa Feevale University https://orcid.org/0000-0001-8107-8675
• Jorge L. V. Barbosa University of Vale do Rio dos Sinos https://orcid.org/0000-0002-0358-2056

DOI: https://doi.org/10.5753/cbie_estendido.2024.243446

Abstract

Affective effects play an essential role in learning, strongly impacting attention and motivation (Tyng et al., 2017). Immordino-Yang and Damásio (2007) highlight the importance of affection in learning processes, especially in relation to students' attention, memory and motivation. This work presents the Uaffect model, which uses affective, educational and personal data of students, based on context histories. The study evaluated the model through a quasi-experimental approach involving 25 high school students. The accuracy results found in the best scenario were 89% for positive affective states and 61% for negative affective states. The tests presented evidence that indicates the possibility of using context histories in the classification and prediction of affective states in educational environments. Thus, Uaffect offers practical contributions to enable teachers to tailor their lessons according to students' emotional states, fostering a personalized educational environment. In addition, the model can serve as a basis for systems that provide realtime feedback, helping teachers adapt their pedagogical strategies. The model also offers significant theoretical contributions to the modeling of systems that use the recognition and prediction of affective states in the educational context. By integrating histories of educational contexts in the identification of affective states, Uaffect substantiates the importance of the learning environment in the emotional experience of students, as highlighted by Moore (2017). This enables a deeper understanding of how contextual factors such as social interaction, physical space, teaching strategies, and methodologies impact students' emotions, thus allowing educators to adjust their pedagogical practices more effectively.

Keywords: Affective Computing, Ubiquitous Computing, Affective States, Context Histories

References

Immordino‐Yang, M. H., & Damasio, A. (2007). We feel, therefore we learn: The relevance of affective and social neuroscience to education. Mind, brain, and education, 1(1), 3-10

Moore, P. (2017). Do We Understand the Relationship between Affective Computing, Emotion and Context-Awareness? Machines, 5(3), p. 16.

Emotion and Context-Awareness? Machines, 5(3), p. 16. Tyng, C. M., Amin, H. U., Saad, M. N., & Malik, A. S. (2017). The influences of emotion on learning and memory. Frontiers in psychology, 8, 235933.