Human Centric Explainable AI for Personalized Educational Chatbots

• Main Authors: H T, Manohara, Gummadi, Annapurna, Santosh, Kathari, Vaitheeshwari, S., Christal Mary, S. Suma, Bala, B Kiran
• Format: Conference Proceeding
• Language: English
• Subjects: Artificial intelligence; Artificial Intelligence (AI); chatbot interactions; Chatbots; Collaboration; Communication systems; Computational modeling; Computer architecture; Educational technology; Explainable AI; Feedback loop; Human-Centric Explainable AI (HCEAI); Learning (artificial intelligence); machine learning; Personalized Educational Chatbots (PECs); Personalized Learning Experiences
• ISSN: 2469-5556
• DOI: 10.1109/ICACCS60874.2024.10716907

With an emphasis on creating individualised learning experiences, the use of artificial intelligence (AI) in educational technology has drawn a lot of attention recently. The Human-Centric Explainable AI (HCEAI) paradigm is presented in this paper to improve the interpretability and transparency of Personalised Educational Chatbots (PECs). The suggested architecture builds an interpretable AI model by fusing sophisticated machine learning techniques with approachable explanations. The HCEAI framework seeks to close the comprehension gap between humans and complicated AI computations by giving priority to the user's cognitive understanding. The chatbot's interactions are enhanced with comprehensible and straightforward explanations through the application of a user-centric design strategy, which promotes engagement and trust. To dynamically adjust the instructional content, the HCEAI system incorporates user preferences, learning styles, and progress data. This tailored strategy improves the chatbot's ability to provide pertinent and interesting learning content. In addition, the system includes feedback loops that let users offer comments on the explanations and help the AI model get better over time. User research was carried out with a group of learners interacting with the personalised educational chatbot in order to validate the efficacy of the HCEAI framework. The outcomes show increased user understanding, contentment, and confidence in the AI model. The framework's flexibility was further demonstrated by its capacity to take into account a range of learning styles. To sum up, this article addresses the need for transparency and user understanding in AI-driven educational technology by presenting a revolutionary Human-Centric Explainable AI framework for Personalised Educational Chatbots. Explainability is integrated to provide a more collaborative and empowered learning environment, while also increasing user trust. The ongoing efforts to make AI a useful and approachable tool for individualised education is aided by this research.