document.ai

Stale
GitHub Python AGPL-3.0

Description

A universal local knowledge base solution based on vector databases and GPT, providing one-stop document processing with vectorization, semantic search, and intelligent Q&A for building private knowledge bases.

Key Features

  • Universal local knowledge base based on vector database and GPT
  • Document upload with vectorization and storage
  • Semantic search and intelligent Q&A
  • Q&A split queries for improved accuracy
  • Topic word extraction to optimize vector similarity
  • Supports custom-trained Embedding models for domain-specific use cases

Use Cases

πŸ’‘ Enterprise internal document knowledge base
πŸ’‘ Intelligent Q&A in customer service chat scenarios
πŸ’‘ Domain-specific Q&A systems for healthcare and other verticals
πŸ’‘ Semantic search over private data
πŸ’‘ Long-term knowledge base maintenance combined with Fine-tune

Quick Start

Clone the repository, install dependencies, prepare a local document dataset, run the vectorization script to store documents in the vector database, start the service, and submit queries via the API to receive intelligent answers.

Related Projects