INODE · Project

Ask Your Databases Questions in Plain English Instead of Writing Code

digitalTestedTRL 6

Imagine you could just ask your company's database a question in plain English — like asking a colleague — instead of needing a programmer to write a query. That's what INODE built: a smart data exploration system that lets anyone search, combine, and visualize information from multiple messy datasets using everyday language. They tested it on real-world data from cancer research, astrophysics, and policy analysis, training actual users to work with it. Think of it as a translator that sits between you and your data, so you don't need to speak "computer" to get answers.

By the numbers

EUR 5,732,000

EU funding for development

consortium partners

countries represented

real-world use cases tested

total project deliverables

user training rounds completed

The business problem

What needed solving

Most companies sit on valuable data they can't fully use because accessing it requires technical skills like SQL, Python, or specialized tools that most employees don't have. When data is spread across multiple systems in different formats — spreadsheets, databases, documents, sensor logs — getting a simple answer often means waiting days for IT or a data analyst to pull a custom report. This bottleneck slows decision-making and leaves insights buried in data that nobody can easily reach.

The solution

What was built

INODE built a suite of data exploration services that let users query multiple datasets using natural language (plain English questions), visual exploration, and query-by-example. The system includes automatic data linking across heterogeneous sources, guided query formulation, and visualization tools. It was validated through 18 deliverables including two rounds of user training on real scientific data from cancer research, astrophysics, and policy analysis.

Audience

Who needs this

Hospital networks and pharma companies managing fragmented clinical and biomarker datasetsBusiness intelligence teams at mid-size companies drowning in disconnected data sourcesOpen data platform operators wanting to make datasets accessible to non-technical usersGovernment agencies and policy units needing quick answers from large public datasetsResearch institutions managing multi-domain scientific databases

Business applications

Who can put this to work

Healthcare & Pharma

enterprise

Target: Hospital networks or pharma companies managing large clinical datasets

If you are a hospital network dealing with mountains of patient records, research papers, and biomarker data scattered across different systems — INODE developed a natural language query system that lets clinicians and researchers search across all these datasets by simply typing questions in plain English. It was specifically tested on cancer biomarker research data at SIB Swiss Institute of Bioinformatics. No SQL expertise needed, no waiting for IT to run reports.

Business Intelligence & Analytics

mid-size

Target: Companies with large heterogeneous data warehouses

If you are a mid-size company struggling to get insights from data locked in spreadsheets, databases, and text documents that don't talk to each other — INODE built tools that automatically link multiple datasets and let non-technical staff explore them through visual analytics and natural language queries. The system was developed with 9 partners across 6 countries and tested on real use cases including policy analysis data.

Research Infrastructure & Data Services

any

Target: Data platforms and open data providers

If you are a data platform provider trying to make your open datasets more accessible to a wider audience — INODE created a suite of services that guides users through data exploration, suggests relevant queries, and visualizes results automatically. Built with EUR 5,732,000 in EU funding and tested across 3 real scientific domains, the technology bridges the gap between complex data and everyday users.

Frequently asked

Quick answers

What would it cost to implement this kind of natural language data query system?

INODE was developed with EUR 5,732,000 in EU funding as a Research and Innovation Action across 9 partners. Licensing or deployment costs would need to be negotiated directly with the consortium. As an open data exploration project, some components may be available as open-source.

Can this scale to handle enterprise-level data volumes?

The system was tested on 3 real-world scientific use cases: cancer biomarker data (SIB Switzerland), astrophysics data (Max Planck Institute Germany), and research policy data (SIRIS Spain). These are large, complex datasets, suggesting the technology can handle substantial data volumes. Enterprise-scale stress testing details would need to be confirmed with the team.

What is the IP situation — can we license this technology?

As an EU-funded RIA project, IP is typically retained by the consortium partners. The coordinator is Zurich University of Applied Sciences (ZHAW) in Switzerland. Licensing arrangements would need to be discussed directly with the relevant partners depending on which components you need.

How mature is this technology — is it ready to deploy?

The project ran from 2019 to 2023 and included user training phases (both initial and final training rounds documented in deliverables). It was tested with real scientific communities, suggesting a working system beyond just a prototype. However, as an RIA, it remains closer to demonstrated technology than a commercial product.

Can it integrate with our existing databases and data systems?

INODE was specifically designed for access-based data integration, meaning it connects to existing datasets rather than requiring data migration. It handles highly heterogeneous data ranging from structured tables to unstructured text, images, and videos. Integration specifics would depend on your current infrastructure.

What kind of support or training would our team need?

The project delivered structured user training in two phases — an initial training round and a final training round — suggesting a defined onboarding process exists. The system was designed for non-technical users, with built-in guidance for understanding data and formulating queries.

Consortium

Who built it

The INODE consortium brings together 9 partners from 6 countries (Switzerland, Germany, Greece, Spain, France, Italy), led by Zurich University of Applied Sciences. The mix is research-heavy: 5 research organizations, 2 universities, and only 2 industry partners (22% industry ratio), with 2 SMEs in the group. This tells a business buyer that the technology is scientifically strong but may need additional commercial packaging before enterprise deployment. The 3 use case providers — SIB (Swiss bioinformatics), SIRIS (Spanish policy analytics), and Max Planck Institute (German astrophysics) — demonstrate the system works across very different data domains, which is a good sign for adaptability.

ZURCHER HOCHSCHULE FUR ANGEWANDTE WISSENSCHAFTENCoordinator · CH
INFILI TECHNOLOGIES SOCIETE ANONYMEparticipant · EL
ATHINA-EREVNITIKO KENTRO KAINOTOMIAS STIS TECHNOLOGIES TIS PLIROFORIAS, TON EPIKOINONION KAI TIS GNOSISparticipant · EL
SIB SWISS INSTITUTE OF BIOINFORMATICSparticipant · CH
LIBERA UNIVERSITA DI BOLZANOparticipant · IT
MAX-PLANCK-GESELLSCHAFT ZUR FORDERUNG DER WISSENSCHAFTEN EVparticipant · DE
SIRIS ACADEMIC SLparticipant · ES
CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE CNRSparticipant · FR

How to reach the team

Zurich University of Applied Sciences (ZHAW), Switzerland — contact via project website or university research office

Order a report on this consortium See ZURCHER HOCHSCHULE FUR ANGEWANDTE WISSENSCHAFTEN profile →

Next steps

Talk to the team behind this work.

Want to explore how INODE's natural language data query technology could work for your organization? SciTransfer can connect you directly with the research team and help assess fit for your specific use case.

Order a report on this topic More in Digital & ICT