Sensor-free Affect Detection in Learning Environments: A Systematic Literature Review
Emotions and affective states influence cognition and learning processes. Computer-based learning environments (CBLEs) capable of detecting and adapting to these states significantly enhance learning outcomes. However, practical constraints often hinder the deployment of sensor-based affect detectio...
| Autores: | , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2024 |
| País: | Brasil |
| Institución: | Sociedade Brasileira de Computação (SBC) |
| Repositorio: | Revista Brasileira de Informática na Educação |
| Idioma: | inglés |
| OAI Identifier: | oai:journals-sol.sbc.org.br:article/4362 |
| Acceso en línea: | https://journals-sol.sbc.org.br/index.php/rbie/article/view/4362 |
| Access Level: | acceso abierto |
| Palabra clave: | Sensor-Free Affect Detection Systematic literature review Emotional Learning Environments Emotion Detection Computer-Based Learning Environments |
| Sumario: | Emotions and affective states influence cognition and learning processes. Computer-based learning environments (CBLEs) capable of detecting and adapting to these states significantly enhance learning outcomes. However, practical constraints often hinder the deployment of sensor-based affect detection in CBLEs, especially for large-scale or long-term use. Consequently, sensor-free affect detection, reliant solely on interaction logs, emerges as a promising alternative. This paper offers a comprehensive literature review on sensor-free affect detection, covering frequently identified affective states, methodologies for sensor development, CBLE attributes, and research trends. Despite the field's maturity, there's ample room for further exploration. Future research should focus on improving sensor-free detection models, collecting more samples of underrepresented emotions, and refining model development practices. Additionally, efforts should be made to integrate models into CBLEs for real-time detection, provide meaningful interventions based on detected emotions, and deepen understanding of emotions' impact on learning. Key suggestions include comparing data collection techniques, optimizing duration granularity, establishing shared databases, and ensuring model source code accessibility. |
|---|