On the Development of a Progressive Web Application Integrating Generative AI and Intelligent Systems for Personalized Learning in Mathematics and STEM Education
Juan David Pérez López, Ómar Ramírez González
Source abstract
This work presents the development of a Progressive Web Application (PWA) designed as a modular intelligent learning environment for personalized education in mathematics and STEM. The system addresses persistent limitations in conventional digital learning platforms, including delayed feedback, weak contextual adaptation, fragmented support between theory and practice, and limited responsiveness to student progression. To overcome these challenges, the platform is built upon a multi-layer architecture integrating a dynamic user model, an adaptive decision engine, telemetry-driven learning analytics, contextual tutoring, and an interactive presentation layer referred to as Interactive Learning Graphic Motion (ILGM), which governs how content is structured, visualized, and activated for each learner. The development emphasizes coordinated interoperability across intelligent subsystems rather than isolated AI functionalities: adaptive planning mechanisms regulate sequencing, difficulty, and representation, while agent-based decision-making components introduce agentic capabilities within study sessions, enabling dynamic content generation, contextual adapptation, and role-oriented instructional behaviors. In this framework, large language models provide contextual tutoring, generative explanations, and real-time learning support, complemented by interactive visual modules for guided practice and multimodal content delivery. To ensure pedagogical consistency and reliability, the system incorporates validation and calibration mechanisms that guide the generation and adaptation of exercises, maintaining coherence in difficulty progression, solution quality, and evaluation criteria across learning sessions. Implemented as a PWA, the system ensures cross-device accessibility, low-friction interaction, and session continuity aligned with modern student behaviors. This paper aims to explain generative artificial intelligence—particularly large language models—and intelligent adaptive systems applied on an online platform can converge within a unified architecture to enhance personalization, interactivity, and continuity in learning processes across mathematics and STEM education.
Evidence graph
No public relationships recorded yet.
Integrity note: This page is a factual metadata record created by deterministic ingestion. It is not a claim that the work moves a mathematical frontier or has been independently verified.