OK, time to put our Cognitive Digital Twin in action. We suppose that we have been able to create our CDT. The Digital Reality Initiative has developed a tool, KaaS, Knowledge as a Service, that supports the creation of your CDT, with knowledge items defined in the IEEE ontology, over 11,000 today – you may want to try it, it has not been released to the public but it can be used as a prototype. Just send me a message with your email and I will ask the developers to provide an access for you.
Using this tool it is possible to access to the huge knowledge base of IEEE, in a way it is the IEEE CDT. Its knowledge is structured according to the IEEE ontology and it provides access to all papers stored in the IEEE Xplore.
Additionally, it contains the link to all conferences and publications, to all societies and education material “owned” by the IEEE. One can browse the knowledge space through keyword, like “bring me in the 5G space”. That will result in the visualisation of the 5G knowledge entity and will be showing the knowledge entities directly connected to it (or those indirectly connected via another knowledge entity – this is what is called a level two map, and it gets very big, possibly too big to be meaningful, so in general one may want to stick with the representation of level 1 knowledge entities, those that are directly connected to the one selected). It becomes possible to find the conferences that might be of interest, because they are dealing with the knowledge entity selected, or the training material existing on that knowledge entity.
So, let’s suppose that using KaaS we have created our CDT as a subset of the IEEE CDT. It will mirror the set of our knowledge (in the IEEE space). Does it mean that it understands them in the same way we do? No, it is an artefact (at least up to now, like a computer it can process data without necessarily having to know what is their meaning: a natural language translator an use algorithms that provide a perfect translation of English into Chinese without having to understand, in our human sense of understanding, either languages). It “knows” that 5G is related to a “wireless system” and can associate the two when needed. It also knows that if there is a paper mentioning 5G and that paper is published today, it might well contain information that updates the one we know (by looking at what we know about 5G it can derive much more … to the point of informing us that there is something new boiling up). It will also embed our knowledge “thread”, that is the way we acquired our knowledge (at school, through training courses, by attending a conference, by reading certain articles, by the kind of work and activity we have done and are doing,…).
We can use our CDT to explain. by proxy, what is our level of knowledge, we could use it as a filter when we do some search on the web to allows us to focus on what we are interested in as well as to grab that information that we are most likely to understand.
Actually, we can do much, much more, but to discover what can be the potential of a CDT we need to understand its evolution. and I will address that in the next post.