Generación de un Modelo de Reconocimiento Automático de Patrones de Actividad Docente en Educación a Distancia: Evaluación del Desempeño y su Impacto en el Aprendizaj

Abstract

Abstract: This research is focused on the development of an artificial intelligence model aimed at achieving an automated pattern recognition system for teaching activity in distance education environments. The model is based on data obtained from the Universidad Técnica Particular de Loja, utilizing advanced artificial intelligence techniques such as Natural Language Processing (NLP) supported by the BERT model, as well as machine learning algorithms to analyze data generated on platforms such as CANVAS, ZOOM, and Gradebook. This approach enables the identification of patterns that can correlate teaching activity with student academic performance. The core of the research lies in the ability to design and train predictive models using neural networks, as well as classification and regression algorithms, subsequently improving their performance through hyperparameter tuning and cross-validation. These models not only automate the analysis of large volumes of data but also provide relevant predictions regarding the existence of relationships between teaching practices and student outcomes, offering the possibility of implementing data-driven interventions. The resulting model enhances educational management by producing tools that enable the objective identification of teaching performance, along with a qualitative assessment of distance teaching interventions. This facilitates the definition of personalized teaching strategies, thereby contributing to the transformation of distance education through more adaptive, equitable, and effective learning approaches.