Cognitive Computing

Why Trust Techopedia

What Does Cognitive Computing Mean?

Cognitive computing describes technologies that are based on the scientific principles behind artificial intelligence and signal processing, encompassing machine self-learning, human-computer interaction, natural language processing, data mining and more. Its aim is to solve complex problems characterized by uncertainty and ambiguity, which in other words means problems that are only solved by human cognitive thought.


Techopedia Explains Cognitive Computing

Cognitive computing is the branch of computer science concerned with solving complex problems that may have dynamically shifting situations and information-rich data that tend to frequently change and sometimes even conflict with each other. A human may deal with such problems by evolving goals and changing objectives, but traditional computing algorithms are not able to adapt to such change. In order to deal with these sorts of problems, cognitive computing systems have to weigh the conflicting data and suggest an answer that best fits the situation rather than what is “right.”

Though there is currently no agreed-upon definition of cognitive computing in the industry or the academe, the term is often used to describe new technology that mimics the way that the human brain functions and how it approaches problem solving. It can be seen as a field that has a goal of accurately modeling how the human mind senses, reasons and responds to stimuli around it. Its greatest applications would be in data analysis and adaptive output, adjusting output to fit a particular audience.

Properties of a cognitive computing system include:

  • Contextual – Understands and extracts contextual elements such as meaning, time, location, process and others based on multiple sources of information. For example, it may be fed with data such as road, ambulance, injury and wreckage and come up with the context of a vehicular accident.
  • Adaptive – This is the learning portion. It adapts to new information and stimuli to resolve ambiguity and tolerate unpredictability. In relation to context, this characteristic takes care of feeding on dynamic data and then processing it in order to form the eventual context and come up with solutions or conclusions.
  • Interactive – The system is able to interact with users so that the users can define their needs, as well as connect with other devices and systems.
  • Iterative and stateful – The systems must aid in the definition of the problem by asking the right questions and finding additional sources of information if a problem is incomplete or ambiguous. They must also be able to remember previous interactions and processes and return to the state at previous points in time.

Related Terms

Margaret Rouse
Technology Expert
Margaret Rouse
Technology Expert

Margaret is an award-winning technical writer and teacher known for her ability to explain complex technical subjects to a non-technical business audience. Over the past twenty years, her IT definitions have been published by Que in an encyclopedia of technology terms and cited in articles by the New York Times, Time Magazine, USA Today, ZDNet, PC Magazine, and Discovery Magazine. She joined Techopedia in 2011. Margaret's idea of a fun day is helping IT and business professionals learn to speak each other’s highly specialized languages.