Трирівнева архітектура інтелектуального telegram бота для персоналізованого навчання

Abstract

The paper addresses the problem of creating a universal intelligent tool for personalized learning in the context of the digital transformation of education. Traditional learning management systems have limitations related to accessibility, adaptability, and integration into the user’s everyday digital environment. To address these issues, an adaptive learning system implemented as a Telegram chatbot using large language models is proposed. The architecture follows a three-layer approach including a user interaction layer, a scenario management layer, and an intelligent content generation layer. The system is implemented using the event-driven aiogram framework, a finite state machine for managing learning scenarios, and a unified API interface for interaction with AI models. The proposed approach enables automatic course structure generation, adaptive content delivery, and monitoring of user learning progress, demonstrating the feasibility of scalable intelligent learning environments integrated into messenger platforms