5 Things To Know About Cognitive Systems
LindaMay 12000078DK Visits (4608)
The growth in big data is increasing dramatically as more and more activity is represented digitally. Now data comes in both structured (identifiable because it organized into a format) and unstructured forms such as video, images, symbols, and natural language. Big data is forcing IT to look for a new computing model in order to help organizations and their decision makers take advantage of all the information available.
Here are 5 things to know:
1. Cognitive systems are...
According to IBM Research, cognitive computing systems learn and interact naturally with people to extend what either man or machine could do on their own. They help human experts make better decisions by penetrating the complexity of Big Data. Think of cognitive systems as having the ability to collaborate with subject matter experts by providing them (experts) with recommendations based on evidence collected from a wide variety of available data (big data).
2. Cognitive systems can supplement human expertise
What is really interesting about cognitive systems is they can handle natural language and analyze the structure of a sentence to determine the underlying intent of the speaker. It can collaborate with its human operator in order to refine its recommendation and eliminate irrelevant information. Cognitive systems take computing to the next level because the exchange between the human operator and the system is more like two experts interacting. Each entity (human operator and cognitive system) has their own experiences and can “discuss” the next course of action for the problem at hand. The cognitive system brings to the exchange a vast and expanding knowledge base while the human counterpart brings their expertise, intuitive sense, and creativity.
3. A cognitive system has an ecosystem to increase its knowledge base
A cognitive system has its own knowledge base derived from human expertise and information. To remain relevant it must be part of an ecosystem that continuously feeds it with new knowledge, information, and expertise. The knowledge fed into the knowledge base can come from experts in the “owning” organization, from broader industry expertise, and from partnering with a consortium that supports the field/domain.
4. A cognitive system’s ecosystem consists of five subsystems
To better understand a cognitive system’s ecosystem lets look at the core subsystems:
5. The cognitive system’s ecosystem should be developed incrementally
How do your create a cognitive system’s ecosystem? Start by using existing systems as the baseline. Existing systems provide the initial information sources for the data fabric. Over time, various essential analytics solutions are deployed in the environment. These solutions deliver value and contribute to the evolution of the ecosystem. The overall ecosystem is expanded and enhanced by focusing on these three aspects in each solution:
To really make the cognitive system robust the overall ecosystem must incorporate incoming information, analytic models, and subject matter experts’ knowledge. As time and these items are applied to nurture the cognitive system’s knowledge base and expand its expertise – it increases in value to the organization.
For more information on cognitive systems, see the IBM Redguide Smar
LindaMay Patterson is an IBM Redbooks Technical Writer. She works with thought leaders from across IBM to create books, papers, blog posts, and videos on cutting edge technology.