The new landscape of knowledge, characterised by:
– continuously expanding data and metadata, partly fuelled by humans and partly by machines, working asynchronously and only loosely connected among them,
– a tremendous ease to access data and at the same time an impossible task to reach all potentially relevant data (think about the easiness of googling information and the impossibility to look at all of the answers provided!) and
– the fact that the access to knowledge can, and most of the time do, increase knowledge
creates a perfect storm.
An interesting approach to tackle the new landscape (not necessarily to solve all the associated problems) is by using machines (soft machines) and connecting them, seamlessly, to organisations and single persons through cognitive digital twins.
We already have machines for searching the web, the proprietary –enterprise and personal- webs and the dark web. These tools, most of the time, open the door to a (large) subset of data that can be potentially relevant to the needs at hand. The visibility afforded is usually skewed to facilitate skimming of what can answer a need. This is based on automated profiling (complemented, in a few occasions by filtering requested by the user). A cognitive digital twin is a tool that can boost current profiling and act as an intermediator of knowledge, in both direction.
Let’s consider first a Digital Twin. Invented some 15 years ago they have taken full shape in this decade and in the last five years have found application in manufacturing and operation. Now they are rapidly expanding in other areas.
Firstly, a digital twin is a digital model of an entity, be it a device, an engine, a product, a process, an organisation. It can represent a simple entity like a hydraulic pump or a complex aggregation of entities (hard and soft) like a building or a city.
Secondly, a digital twin shadows the real entity, being connected to the real entity in a variety of ways that allow the digital model to remain in synch with the real entity (e.g. through sensors picking up the changes).
Thirdly, a digital twin contains the digital thread, basically the history of the evolution of the real entity over time. This digital thread can be used for reverse engineering of a situation, as an example to understand when something happened to affect the current behaviour (an excessive stress on a component might lead to a future malfunction).
As said, a digital twin can be applied to a variety of entities, so why not apply the concept of digital twin to a person, to an organisation? Indeed, we are seeing the first attempts in this direction. As an example a digital twin may become very useful in healthcare, mirroring some genetic characteristics of a person, shadowing its situation (by monitoring health parameters, like heart beats, breathing, blood pressure, glucose, metabolism…) and its evolution over time (what has been eaten, the kind of physical exercise taken, medical exams and drugs….). It can also be applied to a community detecting the overall health status, the presence of pollutants, monitoring the upstart of epidemics, the effect of prophylactic measures.
The next step, that is the one I am interested in this series of post, is the possibility to apply the concept of digital twin to knowledge, to create a cognitive digital twin serving both a person and an organisation.