Proof-of-Concept Pilot This is an early preview still in testing. Results may be incomplete, inaccurate, or contain mistakes. Your feedback is very welcome.

Matchmaker

Search by research topic, theme, or a researcher's name. The tool uses AI to find the most relevant THUAS publications and the people behind them, and it tolerates small typos in names.

Search:
Try a topic:
Or a name:

Ask

Ask a research question in your own words. The assistant calls the matchmaker for you and answers with researchers and publications cited from the THUAS index.

Knowledge Graph

Interactive research landscape — Domain → Field → Subfield → Topic → Publication → Researcher

Click to expand: domain → field → subfield → topic → publication → author → co-authors
×
Domain
Field
Subfield
Topic
Publication
Research Group

Research Explorer

Browse the CERIF research database — publications, researchers, topics, organisations & SQL

Dashboard Publications Researchers Topics Organisations SQL
TitleYearTypeLanguage
NamePublications
KeywordPublications
NameTypeResearchers
Ctrl+Enter
Example Queries
Schema

About This Pilot

Why we built it, what it does today, and where it is going

Please note: This is a proof-of-concept pilot. The underlying data, classifications, and search results are still being tested and refined. You may encounter gaps, incorrect matches, or missing researchers. Treat the results as a starting point for discovery, not as a definitive record.

Why We Built This

The Hague University of Applied Sciences (THUAS) conducts a wide range of research across many faculties and research groups. Finding out who is working on what, and discovering potential collaborators inside the university, is harder than it should be. Publications are spread across repositories, researcher profiles are incomplete, and there is no single place to explore what the university is producing.

This pilot investigates whether a matchmaking tool can help. By combining open publication data, AI-based topic classification, and semantic search, the tool aims to make THUAS research easier to find, easier to connect, and easier to build upon, for staff, students, and external partners.

The project is commissioned by the AI Hub and developed by the AI & Data Science Expert Team at THUAS.

What the Tool Uses

SURF Sharekit

The national Dutch research repository. All THUAS publications registered in SURF Sharekit are harvested as the starting dataset (~4,400 records).

OpenAlex

An open catalogue of scholarly works that provides a standardised topic taxonomy used worldwide. Each publication is linked to topics from this taxonomy.

LLM Classification

A language model assists in reviewing each publication and assigning relevant topics and keywords, helping to enrich records that have limited metadata.

Semantic Search

Queries and topics are encoded as vectors so the tool can match meaning rather than exact words. For example, “ageing population” finds work on “elderly care”.

Research Database

A structured database of publications, researchers, research groups and faculties, following the European CERIF standard for research information.

Knowledge Graph

An interactive map showing how research domains, topics, publications and researchers are connected across the university.

What You Can Do Today

  • Ask in plain language: open the Ask tab and type a question like “Who at THUAS works on AI in healthcare?”. An AI assistant searches the THUAS index for you and answers with researcher names and publications cited. External users can plug in their own OpenAI or Azure OpenAI key from the panel; the key stays in your browser.
  • Search by topic: in the Matchmaker, describe a research area (for example, “AI in healthcare” or “sustainability in urban environments”) and discover the most relevant publications and the researchers behind them.
  • Search by name: type a researcher's name (typos and partial names are tolerated) to find their publications and related people.
  • Mixed queries: combine the two, for example “Bolte smart sensor systems”. The tool detects the name and the topic separately and highlights papers that satisfy both.
  • Auto / Topic / Person modes: a small toggle next to the search bar lets you override the auto-detection.
  • Paper details: click any publication's title (or the “View details” button) to see authors, DOI, the full topic hierarchy, keywords, and a Google Scholar link as a fallback.
  • Knowledge Graph: browse the THUAS research landscape from broad domains down to individual publications and researchers.
  • Research Explorer: browse the structured database of publications, researchers, topics, and organisational units.

What Is Next

Person search and the conversational Ask tab are now live. The next big steps are education-data integration, adding missind research data, dedicated profile pages for each person, and a formal evaluation of how well the matchmaker surfaces the right results. Note that these are not in any order of priority.

Next

Researcher Profile Pages

Click a researcher to see a dedicated page with their full publication list, topics they work on, collaborators, and faculty/research-group affiliations and education data.

Planned

Education Integration

Connect research output to teaching so you can see which courses or curricula relate to a given research area and more.

Planned

Open API for AI clients

Publish an OpenAPI spec so anyone can plug the matchmaker into ChatGPT Custom GPTs, NebulaONE agents, Claude, or other AI clients without using our chat tab.

Planned

Evaluation & Quality

Formal evaluation of how well the matchmaker surfaces the right results, with input from researchers across the university.

Feedback Welcome

Because this is a pilot, your feedback is very valuable. If a search returns unexpected results, if your own publications appear incorrectly classified, or if someone you would expect to find is missing, please email, Amey Vasulkar at anvasulkar@hhs.nl or get in touch with the any AI Expert Team member at THUAS. Each piece of feedback helps us improve the data, the matching, and the tool itself.