Advancing Learning Analytics with Machine Learning and Sentiment Analysis

The recent research in the field of educational technology and learning analytics has seen significant advancements aimed at enhancing student engagement and improving teaching strategies. A common theme across the studies is the use of machine learning and data analytics to interpret and predict engagement patterns, leveraging large datasets from virtual learning environments and audio-recorded lessons. Innovations include the development of refined engagement metrics, semi-automated analysis methods for teacher discourse, and sentiment-based engagement metrics using large language models. These approaches not only optimize the time and resources required for analysis but also provide scalable and accurate measures of engagement, enabling early intervention strategies and data-driven teaching practices. Notably, the integration of large language models for sentiment analysis and discourse classification has shown promising results in enhancing the precision and efficiency of engagement assessment. These developments collectively push the boundaries of learning analytics, offering new tools and insights for educators and researchers to better understand and support student engagement.

Sources

Uncovering Student Engagement Patterns in Moodle with Interpretable Machine Learning

Semi-automated analysis of audio-recorded lessons: The case of teachers' engaging messages

Enhancing Talk Moves Analysis in Mathematics Tutoring through Classroom Teaching Discourse

LLM-SEM: A Sentiment-Based Student Engagement Metric Using LLMS for E-Learning Platforms

Towards an optimised evaluation of teachers' discourse: The case of engaging messages

Built with on top of