Discover and explore top open-source AI tools and projects—updated daily.
Chatbot for zero-shot information extraction research
Top 43.0% on SourcePulse
ChatIE is an open-source tool for zero-shot information extraction (IE) using large language models, specifically ChatGPT. It targets researchers and developers needing to extract structured data like entities, relations, and events from unannotated text with minimal human intervention, offering a cost-effective alternative to traditional supervised IE methods.
How It Works
ChatIE transforms zero-shot IE into a multi-turn question-answering problem within a two-stage framework. By leveraging ChatGPT's capabilities and carefully crafted prompts, it directly extracts structured information from raw sentences. This approach bypasses the need for extensive data labeling, enabling rapid development and deployment of IE systems, even for complex tasks like entity-relation triple extraction, named entity recognition, and event extraction.
Quick Start & Requirements
front-end
and run npm install
then npm run start
. Navigate to back-end
and run python run.py
.Highlighted Details
Maintenance & Community
The project is associated with the paper "Zero-Shot Information Extraction via Chatting with ChatGPT" (arXiv:2302.10205). The README indicates active updates are planned. No specific community channels (Discord/Slack) are listed.
Licensing & Compatibility
The repository is presented as open-source, but the specific license is not explicitly stated in the README. The use of OpenAI's API implies adherence to their terms of service.
Limitations & Caveats
The online demo is temporarily unavailable due to cost. The response speed and reliability are dependent on the official OpenAI ChatGPT API, which can be slow or overloaded. Using the default OpenAI key is discouraged due to potential overload from multiple users.
1 year ago
Inactive