Research IT

Introducing Ontology Maker: A TIC Initiative Bridging Language Gaps in Disaster Relief

In collaboration with Visioning Lab, the Research Software Engineering team has developed Ontology Maker, an AI-powered platform that helps international teams manage key terminology across languages and disciplines. Initially focused on the temporary architecture sector, the tool simplifies the creation of multilingual glossaries to support faster, more effective disaster response.

Developed as a Collaborative Research & Development (CR&D) project under the Turing Innovation Catalyst, Ontology Maker enables professionals from diverse backgrounds to align their communication. By improving coordination across projects, teams, and services, it enhances the impact of humanitarian efforts around the world. Jessica Symons (Director, Visioning Lab) , Mengmeng Tan ( Postdoctoral Research Associate), and Erdem Atbas (Research Software Engineer) summarise the key features and benefits of this innovative AI application.

Why It Matters

In disaster zones, clear communication can mean the difference between life and death. Yet teams often struggle to align on language, whether due to linguistic differences or sector-specific jargon. Ontology Maker addresses this gap by making critical terms understandable across cultures and professions. Whether designing temporary shelters, managing supplies, or speaking to local communities in the field, the tool ensures everyone shares a common understanding. By turning complex terminology into a shared resource, Ontology Maker supports faster response, smoother collaboration, and stronger recovery efforts around the world.

A First Look at Ontology Maker

Ontology Maker is built to support real-world needs in high-stakes situations. The current prototype offers the following features:

  • Multilingual Glossary Generation: Instantly generate a list of disaster-response tools searchable in English, Nepali, or Chinese, and download the glossary for easy logistics planning and field use.
  • Document-to-Glossary Conversion: Upload a Word, PDF, or TXT file to transform it into a well-structured academic glossary of around 100 key terms and definitions, using your OpenAI API key.
  • Audio Transcription for Field Use: Upload an M4A audio file (such as an interview or voice briefing), and Ontology Maker will transcribe it into clean, editable, and searchable text, making it easier to quote, archive, and share field content.

Ontology Maker Deployment for Public Accessibility

Under the guidance of Erdem Atbas, Research Software Engineer (RSE), the prototype was optimised and deployed to a cloud-hosted virtual machine, making it accessible to a wider audience. The images below highlight the key functionalities of the Ontology Maker.

Ontology Maker 1

Figure 1: This figure depicts glossary searching via Ontology Maker, allowing users to search multiple items in English, Nepali, or Chinese.

Ontology Maker 2

Figure 2: This figure illustrates the Automated Glossary Extractor, enabling users to create their own glossary, with structured output generated using ChatGPT.

Ontology Maker 3

Figure 3: Automated Audio Transcription

During the research, a key question emerged: should Ontology Maker include automated audio transcription to allow users to upload audio files (e.g., .M4A) and generate glossaries directly from spoken content? To make this possible, Ontology Maker integrates Whisper, a transcription tool, to convert audio into text.

Future Development and Integration

To further expand the capabilities of Ontology Maker, future developments will include:

  • Advanced Translation Services: Expanded multilingual support to ensure clear, consistent, and culturally accurate communication across even more languages.
  • GIS and Mapping Integration: Integration with mapping tools to deliver location-based terminology insights, supporting spatially-aware glossary use and planning.
  • Automated User Feedback Collection: Real-time feedback features to allow users to suggest changes, improving glossary relevance, clarity, and accuracy through active community input.

These enhancements will support more precise placemaking, enabling teams to define and discuss locations more effectively, which is crucial in both urban recovery and community rebuilding efforts.

Does Your Research Require an AI Software Solution?

If you’re interested in working with the RSE team on an AI-related project (large or small), get in touch with us via Connect and we’ll organise a meeting to discuss how we can work together. Find out how the University of Manchester is driving growth across Greater Manchester via AI collaboration AI Innovation: The Turing Innovation Catalyst's Success