A back arrow icon.
RDFox Blog

Ask Ian: What is KRR?

Ask Ian: What is KRR?
Ian Horrocks

In the first of our ‘Ask Ian’ podcast series, we interviewed one of the co-founders of OST, Ian Horrocks.

Ian is the winner of the 2020 BSc Lovelace Medal for his contributions to the field of Computer Science, so we thought he’d be perfect to tell us all about Knowledge Representation and Reasoning.

To listen along, check out our podcast episode below, available on YouTube.

I’m Ian Horrocks. I’m a professor of Computer Science here at the University of Oxford, and my main area of research is Knowledge Representation and Reasoning.

So, what is KRR?

KRR is knowledge representation and reasoning. Knowledge can be of different kinds. Knowledge can be basic facts like you would find in a database, but knowledge can also be more complicated rule-based knowledge that tries to generalise.

Let’s say for example, in a company setting where they make a large number of components which they combine and configure to build larger installations for clients. The knowledge would consist of basic stuff about the components. If they were electrical components, it would be things like what voltage or power a given motor has.

Then there’s also more complicated knowledge that includes rules of the form: “If I have a motor and I need a power supply, then the motor should take the same voltage as the power supply provides.” — this is so that it’s a good configuration in the end for the client.

The reasoning part is how we combine all that stuff in order to answer questions. A question might be of the form, “Does this configuration, which involves motor A, power supply B, and a tonne of other things, will that solve a given customer requirement?”.

As I’ve said, this is just one example of the use of complex reasoning in a commercial application which allows the system to answer arbitrary questions, while taking into account domain knowledge that’s specific to that application.

The reasoning that’s taking place there can also be thought of as rules-based AI.

What does that reasoning look like?

Actually, one of the important things about ‘Knowledge Representation and Reasoning’ systems is that, as the user answering the question, you don’t have to know how to do the reasoning—that’s the whole point!

You just have to know that some very clever people in the background have figured out a computer algorithm that does the reasoning for you and guarantees to be correct. So, if it gives you an answer, you know that that’s a correct answer, and if a correct answer exists, it will always find it. So, you just rely on the fact that someone else did the heavy lifting on all of that stuff.

It's the result of all of this research and thinking about KRR that ended up in RDFox, the rules-based AI system that I’m going to talk about more in future podcasts.

Stay tuned for future podcast episodes, where we ask Ian more about how RDFox was created and the applications of KRR in various industries.

Check out our other interview series, ‘Meet the Founders’, where we asked our founders about their journey in bringing OST to life:

Take your first steps towards a solution.

Start with a free RDFox demo!

Take your first steps towards a solution.

Get started with RDFox for free!

Team and Resources

The team behind Oxford Semantic Technologies started working on RDFox in 2011 at the Computer Science Department of the University of Oxford with the conviction that flexible and high-performance reasoning was a possibility for data-intensive applications without jeopardising the correctness of the results. RDFox is the first market-ready knowledge graph designed from the ground up with reasoning in mind. Oxford Semantic Technologies is a spin-out of the University of Oxford and is backed by leading investors including Samsung Venture Investment Corporation (SVIC), Oxford Sciences Enterprises (OSE) and Oxford University Innovation (OUI).