An Intelligent AI-Integrated Online Learning Management System

Các tác giả

  • Nam Vũ Hải
  • Thịnh Nguyễn Đức
  • Trung Trần Thế Industrial University of HCMC

Từ khóa:

Online Learning Management, Context-Aware AI, Next.js, Docker, Cloudflare Tunnel, Gemini API, E-learning Platform

Tóm tắt

The integration of AI chatbots into learning management systems presents a significant opportunity to personalize the educational experience. To fully realize this potential, the core challenge of "context-awareness" understanding who the learner is, their progress, and the specific lesson content must be addressed. This paper introduces Cognistream, an intelligent LMS architecture designed to solve this challenge. The core contribution of the system is the "Dynamic Prompt Adaptation" architecture. This architecture utilizes a "Conversation Analyzer" and a "Context Fetcher" module to automatically enrich user queries with real-time academic data, including user progress, lesson content, and video transcripts, before calling the Gemini API. To validate, we deployed the system on a secure, high-performance architecture (NestJS, Next.js, Docker, Cloudflare Tunnel) and conducted an empirical analysis. The research results indicate two key findings: the context-aware AI provides significantly more accurate and personalized responses compared to a generic AI model; and the deployment architecture not only ensures security (origin IP hiding) but also achieves outstanding access performance with an average page load time of 656ms. This research confirms the value of integrating context-aware AI in creating truly personalized learning pathways.

Đã Xuất bản

09-12-2025

Số

Chuyên mục

Kỹ thuật phần mềm (Software Engineering)