extracting "structured" information that is embedded in natural language text on the web using iterative set expansion, spanBERT, and openAI API - View it on GitHub