Aplicación móvil de análisis de sentimientos en tiempo real con IA para medir la interacción en tecnologías de la información.
This research addresses the challenge educators face objectively and in real-time when evaluating student interaction during teaching-learning, especially in contexts involving large groups or fast-paced dynamics. The main objective was to develop a mobile application based on artificial intelligenc...
Salvato in:
| Autore principale: | |
|---|---|
| Natura: | bachelorThesis |
| Lingua: | spa |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | http://dspace.unach.edu.ec/handle/51000/15440 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
Aggiungi Tag | Riassunto: | This research addresses the challenge educators face objectively and in real-time when evaluating student interaction during teaching-learning, especially in contexts involving large groups or fast-paced dynamics. The main objective was to develop a mobile application based on artificial intelligence (AI) capable of performing real-time sentiment analysis through facial recognition to measure student interaction in the Information Technology Engineering program. An applied and experimental methodology was employed, utilizing AI techniques such as OpenCV and DeepFace for emotion recognition. The application was developed using React Native under a microservices architecture. Project management followed the agile Kanban methodology, and the system's usability was evaluated according to the ISO/IEC 25010 standard through surveys administered to instructors. Student interaction was measured by analyzing facial emotions at the beginning and end of each class, classifying detected sentiments such as happiness, sadness, anger, fear, surprise, and neutrality. Based on this data, comparative charts were created to identify emotional variations. Positive emotions (happiness and surprise) were interpreted as indicators of stronger emotional engagement and, therefore, a higher level of interaction, while negative or neutral emotions suggested disinterest or disengagement. This methodology allowed for a direct link between emotional states and active student participation. The results demonstrated that the application could accurately detect predominant emotions and highlight significant changes in interaction levels before and after class. Over 60% of participating instructors strongly agreed with the tool's usefulness, accuracy, and usability. It is concluded that the developed application represents an innovative solution for objectively measuring student participation and provides a valuable tool for educators by offering real-time emotional insights that can enhance pedagogical strategies. |
|---|