Knowledge representation and reasoning
Whenever we want to make a decision or check something, correct information is crucial. The more clear the information, the better. If not, and for example if different terms are used for the same concept, there will be disruptive ‘noise’. Then you won’t be able to guarantee effective decision support or real-time control. The problem, however, is that many systems offer ambiguous information.
To solve this problem, we use the AI technology 'knowledge representation & reasoning'. This enables us to recordknowledge from the domain so thatthe same terms are used consistently and information is presented clearly. Only then can we reason about this knowledge in the right way. New facts and insights can be derived, which strengthens decision support and real-time control.
OWL formats present unambiguous information
TNO brings visibility to the most important concepts and their interrelationships within a domain. We present these knowledge models using various AI tools in the form of Knowledge Graphs of Ontology Web Language (OWL) formats.
We develop such an ontology in collaboration with domain experts. By presenting relevant information and possible trends in an unambiguous way, a reasoning machine can exploit this ready-to-use data to derive new insights and connections.
Knowledge representation & reasoning is a key technology within AI. It supports decision-making and makes real-time control effective. - Jack Verhoosel - senior business consultant & architect
Hybrid AI is twice as powerful
The AI knowledge representation & reasoning technique can be very valuable in combination with machine learning. In this way, machine learning contributes by taking into account the uncertainty of predictions. And knowledge representation & reasoning, in turn, helps to interpret the internal relationships within a domain.
This combination of reasoning (knowledge representation & reasoning) and learning (machine learning) is called Hybrid AI. In addition to decision support and real-time control, this also helps to find trends and patterns. This technology is currently used in various sectors, such as agriculture, horticulture, defence, industry and energy.
Linda is consultant in de department Data Science. Linda develops semantic standards and advises organizations on how to become interoperable. Her specific focus is on how organizations can develop and manage their semantic standards, if the organization is dynamical and thus needs a combination of expert knowledge and data. Typical customers for Linda are branche or standardization organizations. She works in several domains, the last years with a focus on the labour market with projects such as Skills Matching and SETU.
Looking for an expert?View all experts
System integration for robots in greenhouses
Good collaboration between humans and AI robots in greenhouses is essential. We are working on a semantic explanation and navigation system for improved communication.
AutoAdapt: Self-adaptive machine learning to propel us into the future
Self-adaptation in computing is a concept that might prove to be the missing link in making AI more transparent and accelerating innovation.
Developing moral models for AI systems
It is vital that values such as safety, trust and well-being are integrated in the decision-making process. How far are AI systems that we can trust them?
First overview of cyberattack techniques by AI against AI
All international trend reports view Artificial Intelligence as the most important disruptive technology of the coming years. Where a new technology develops, new vulnerabilities also arise. And AI is no exception.
‘Giant AI goes down the European road’
TNO supports the alarming call of the Future of Life Institute regarding AI. Regulation is urgent and cannot be left solely to the market. Read TNO's response.