AI-Enabled Pedagogical Framework for Integrating Indian Knowledge Systems in Teacher Education

Abstract

Indian Knowledge Systems (IKS) encompass a diverse body of intellectual, scientific, philosophical, cultural, linguistic and educational traditions that the National Education Policy (NEP) 2020 seeks to embed in contemporary education. Integrating IKS into teacher education, however, is hindered by fragmented digital resources, heterogeneous knowledge formats, limited personalization, weak pedagogical mapping and the absence of intelligent mechanisms for recommending context-specific resources. This paper proposes an AI-enabled pedagogical framework for integrating IKS into teacher education. The framework combines natural language processing (NLP), semantic analysis, knowledge representation, recommendation techniques, retrieval-augmented generation (RAG), adaptive learning, intelligent assessment and learning analytics. It comprises five layers: Data Collection, AI Processing, Pedagogical Intelligence, Application, and Analytics and Feedback. These layers manage curated IKS resources, perform semantic classification and knowledge-graph construction, map resources to teacher-education learning outcomes, and generate adaptive learning pathways. A prototype dashboard integrates an IKS content library, a source-grounded AI tutor, an adaptive learning module, an assessment engine, a recommendation system and a learning analytics view covering areas such as Indian mathematics, Ayurveda, Yoga, philosophy, astronomy, agriculture, architecture, linguistics, arts and traditional pedagogy. Evaluation of recommendation accuracy, learning improvement, user satisfaction, assessment performance and usability is outlined. The framework offers a technology-supported means of contextualising IKS within teacher education while retaining human oversight over interpretation and knowledge validation, positioning AI as an enabler rather than a replacement for teacher educators and domain experts.