Markup is a web-based document annotation tool designed for Natural Language Processing (NLP) and Machine Learning (ML) tasks, such as named-entity recognition. It targets ML/NLP practitioners and researchers by streamlining the annotation process through AI-powered predictions and integrated ontology mapping. The tool aims to increase annotation efficiency and accuracy by learning from user input and suggesting complex annotations and concept mappings.

How It Works

Markup leverages a machine learning model, powered by GPT-4, to provide predictive annotations and ontology mappings. As users annotate documents, the system learns their patterns and suggests relevant tags or concepts. It integrates with common ontologies like UMLS and SNOMED-CT, and supports custom ontologies, enabling users to map annotated text to standardized terminologies. This approach aims to reduce manual effort and improve consistency in data labeling.

Quick Start & Requirements

• Install dependencies: yarn install
• Start Supabase: supabase start
• Add Supabase API URL/key and OpenAI API key to .env.local.
• Run development server: yarn start
• Access at http://localhost:3000.
• Prerequisites: Node.js, Yarn, Supabase CLI, OpenAI API key.
• Documentation: Quick Start Guide

Highlighted Details

• Predictive annotation powered by GPT-4.
• Integrated access to common (UMLS, SNOMED-CT, ICD-10) and custom ontologies.
• Predictive ontology mapping for concept mapping.
• User-friendly interface for both technical and non-technical users.

Maintenance & Community

Contact: sam@getmarkup.com

Licensing & Compatibility

The repository does not explicitly state a license.

Limitations & Caveats

Requires an OpenAI API key for full functionality, which may incur costs. The project appears to be primarily maintained by a single individual, indicating a potential bus factor.