A Chatbot for Data Management Course Based on Rasa Framework

The integration of technology into education has transformed how students engage with course content. Consequently, the development of educational chatbots is an innovative breakthrough that has been investigated due to its ability to offer learners on-demand support. This study aims to develop and evaluate an Artificial Intelligence (AI) bot called Data Management Course Bot (DMCBOT) to help students in a Scientific Data Management (SDM) course using the opensource machine learning Rasa framework. To improve intent classification accuracy, a custom Support Vector Machine (SVM) classifier replaces standard classifiers by using domainspecific features, vocabulary, and context. A dedicated website was developed as the interface for the Rasa-trained model. Performance evaluation shows DMCBOT achieves 85.4% accuracy, 0.87 precision, 0.93 recall, and 0.90 F1-score, demonstrating strong educational support.