Skip to content

Tuprott991/tuvandaihoc

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

University Advisory AI Project

Project Overview

This project creates an AI-powered university advisory system for Vietnamese higher education institutions. It leverages web crawling technologies to collect comprehensive data from universities nationwide, processes this information, and offers personalized educational guidance through an interactive AI agent.

Core Technologies

Data Collection

  • Crawl4AI: Extracts data from various university websites across Vietnam
  • Deep crawling capabilities for targeted university sites (e.g., hcmus.edu.vn)
  • Markdown processing for structured data preparation

Backend (Python - FastAPI)

  • Database: PostgreSQL hosted on cloud platforms (Azure/AWS)
  • AI Components:
    • Gwen 2 8B model fine-tuned with QLora
    • vLLM deployment on GPU server
    • RAG implementation (custom or lightRAG)
    • LangChain for agent construction
    • Crawl4AI integration with Serper for data discovery and extraction

Frontend (TypeScript)

  • React as the main framework
  • ChakraUI for component styling
  • Key Pages:
    • Homepage
    • Hotline contact
    • University tags/categories (HCMC focus)
    • About us
    • Chatbot interface

Team Structure

Backend Team

  • Database architecture and management
  • LangChain agent development
  • RAG system implementation
  • Web crawling and data processing

Frontend Team

  • Core UI framework development
  • Page implementation and user experience design
  • Chatbot interface integration

Features

The AI agent functions as a university tour guide, capable of:

  • Recommending suitable universities and majors
  • Providing recent admission scores
  • Assessing student capabilities against program requirements
  • Answering detailed questions about specific institutions

Deployment

The system uses GPU-accelerated servers for model hosting with vLLM for efficient inference and API access via FastAPI endpoints.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •