Entity Extraction with LangExtract

This project demonstrates how to use LangExtract with OpenAI models to extract structured information such as person names, countries, and skill sets from unstructured text.

How It Works

1. Define extraction examples (EXAMPLES) and a prompt (PROMPT) for guidance.
2. Pass input text to the extract_result function.
3. Extracted entities (person_name, country, skill-set) are returned with attributes and character intervals.

Sample Output

person_name Amir {'country': 'Egypt'}
country Cairo {'person_name': 'Amir'}
skill-set sketching intricate blueprints {'person_name': 'Amir'}

Installation

pip install -r requirements.txt

Requirements

• Python 3.8+
• LangExtract
• OpenAI API key set as environment variable:

export OPENAI_API_KEY="your_api_key_here"