Document Cracker is an AI-powered document search and analysis tool developed by Global Affairs Canada to help government officials quickly find and understand information across large volumes of documents. The system uses advanced machine learning and natural language processing to automatically index documents, monitor for emerging trends and topics, and identify mentions of key people, locations, and organizations. This enables officials to rapidly identify pressing international issues, develop informed positions, and track how other countries' positions are evolving.
The system is currently in production and is used by Government of Canada employees. It does not involve the collection or processing of personal information. Document Cracker processes unclassified internal GAC documents such as situation reports, meeting readouts, and official statements, along with external news sources. The system leverages Azure Cognitive Search, machine learning algorithms, and GPT-3 for document summarization to provide comprehensive analysis capabilities.
