Four Steps to Landing an Education Scene
1. Automated integration of teaching resources
Creating a Subject Knowledge Crawler:
python cli.py --question "收集初中物理力学相关实验视频" --scraper-provider firecrawl --output-format json --save-path physics_resources.json
2. Intelligent Q&A system construction
- Configuration of education-specific model chains::
existlitellm_config.yamlAdd in:- model_name: edu-helper litellm_params: model: deepseek-teacher temperature: 0.3
- Deploying the Web Interface: Modify the front end to a simplified version suitable for students
3. Learning behavior analysis
Log data is processed through asynchronous processing:
- configure
ANALYTICS_MODE=TRUE - Use the preset analysis templates:
python analytics.py --template learning_pattern
4. Recommendations for customized implementation
- Primary level: enabled
SAFE_SEARCH=STRICTfiltering mechanism - Higher education: connecting to the school library API (
LIBRARY_API_KEY) - Teacher's side: add
--pedagogicalParameter Generation Lesson Plan
Typical cases show that the system can reduce teachers' material preparation time by 60% and enhance students' independent learning efficiency by 2.3 times.
This answer comes from the articleII-Researcher: Deep Search and Stepwise Reasoning to Answer Complex QuestionsThe
May not be reproduced without permission:AI productivity tools " How to apply II-Researcher in educational scenarios to build intelligent learning assistants?
English 
简体中文 
日本語 
Deutsch 
Português do Brasil