What is DeepTutor
DeepTutor is an AI learning assistant open-sourced by the HKU Data Science Lab. Through a multi-agent architecture and knowledge graph technology, it transforms complex knowledge systems into an efficient learning experience. Users upload various documents, and the system quickly builds a knowledge base, using intuitive visualization and interactive learning methods to help users deeply understand knowledge. DeepTutor accurately provides personalized learning paths, greatly shortening learning time and improving knowledge retention, becoming an efficient knowledge accelerator for learners and programmers.
Main features of DeepTutor
-
Large-scale document knowledge Q&A: Supports uploading textbooks, papers, manuals and other documents to build an AI knowledge base, combining RAG, web search and code execution to provide accurate answers and cited sources.
-
Interactive learning visualization: Transforms complex concepts into intuitive visual aids, helping users better understand and remember knowledge through step-by-step analysis and dynamic demonstrations.
-
Intelligent exercise generation: Generates targeted practice questions and mock exams based on the user's knowledge level and learning goals, and supports uploading reference exam papers to match the style of real exams.
-
Deep research and creative generation: Through multi-agent collaboration, conducts systematic research on complex topics, generates research reports and novel research directions, and supports cross-disciplinary knowledge integration.
-
Personal knowledge base management: Users can create and manage their own knowledge bases, with support for incremental updates and cross-document knowledge associations, facilitating long-term learning and review.
How to use DeepTutor
-
Environment setup: Clone DeepTutor's GitHub repository https://github.com/HKUDS/DeepTutor到本地，安装所需的依赖。推荐使用
condaorvenvcreate a Python environment to ensure the system runs smoothly. -
Configure environment variables: Copy the
.env.examplefile as.env, and fill in the necessary API keys (such as the OpenAI API key) so that the system can call external services. -
Launch the application: Choose quick deployment via Docker, or start it manually.
-
Create a knowledge base: Open DeepTutor's web interface, click "New Knowledge Base", and upload the documents you need to study (such as PDF, TXT, Markdown and other formats). The system will automatically process the documents and build the knowledge base.
-
Start learning: In the web interface, obtain knowledge answers through intelligent Q&A, generate targeted practice questions, conduct deep research, or gain a deeper understanding of complex concepts through interactive visualization.
DeepTutor project address
- Project website：https://hkuds.github.io/DeepTutor/
- GitHub repository：https://github.com/HKUDS/DeepTutor
Application scenarios of DeepTutor
-
Technical learning and programming education: DeepTutor helps programmers quickly master new technologies, understand algorithm principles, and prepare for technical interviews, improving learning efficiency through intelligent Q&A, visual explanations and exercise generation.
-
Academic research and paper writing: Researchers use DeepTutor for literature reviews, research idea generation and paper writing assistance, quickly sorting through literature, discovering research gaps and writing high-quality papers.
-
Online education and personalized learning: DeepTutor provides personalized learning paths, interactive learning content, and language learning assistance for online education, enhancing students' learning experience and improving learning outcomes.
-
Enterprise internal training: Enterprises use DeepTutor to build internal knowledge bases for new employee training and employee skill improvement, helping employees quickly become familiar with technologies and business processes.
-
Personal knowledge management: Individual users use DeepTutor to organize learning materials, build knowledge bases, achieve cross-domain knowledge integration and efficient retrieval, and improve knowledge management capabilities.