OntoScope

OntoScope is a web-based system prototype that implements a conceptual interaction model for LLM-based ontology scoping with competency questions (CQs), accessible at ontoscope.digital. It supports ontology engineers in progressively defining a well-scoped ontology by organising LLM-generated CQ candidates spatially along two dimensions: subdomains (horizontal) and term granularity (vertical), enabling both divergent thinking (exploring and generating new CQ candidates) and convergent thinking (refining and eliminating existing ones).

OntoScope_Preview.mp4

Installation

Prerequisites

• Node.js 20.x or higher
• npm (comes with Node.js)
• An OpenAI API key

Steps

1. Clone the repository

   git clone https://github.com/King-s-Knowledge-Graph-Lab/OntoScope.git
   cd ontoscope

2. Install dependencies

   npm install

3. Start the application

   npm run dev

4. Access the application

   Open your browser and navigate to http://localhost:5000

5. Enter your OpenAI API key

   When prompted, enter your OpenAI API key to enable AI-powered CQ generation.

Usage

1. Enter a target domain (e.g., "University Administration", "Healthcare Informatics")
2. Click "Generate Space" to initialize the CQ space with AI-generated subdomains and granularity levels
3. Click on any intersection to explore and add competency questions
4. Use the visualization to identify gaps, delete redundant CQs, and refine terminology
5. Export your finalized CQ space as JSON for use in ontology development tools