Cognitive system

from Wikipedia, the free encyclopedia

A cognitive system is a digital system with interfaces between the digital world and the environment that can perceive and understand things and draw conclusions and learn from them. Cognitive systems are able to independently develop solutions for human tasks. They can interact and cooperate with other digital systems, interpret contexts and are adaptable.

functionality

Example scheme of a cognitive system

Signals from the environment can be detected with sensors such as cameras , microphones or temperature sensors. For perception , these signals are recognized by the system's cognition and converted into digital information. This information can be documented and is processed . The result of the deliberation can also be documented and is used to control and execute ("execution") an action in the environment with the help of actuators .

If there is feedback between the actuators and the sensors, the cognitive system can form a digital control loop .

Typically , cognitive systems communicate with other systems in order to exchange , verify or validate data . Through cooperation , several systems can work simultaneously in the digital and real world. Humans can interact with cognitive systems via human-machine interfaces .

For the implementation of cognitive systems, for example, neural networks , big data or machine learning or deep learning are used.

Application examples

Cognitive systems are used in an increasing number of areas and represent, for example, the basic technology for self-driving vehicles , intelligent personal assistants , Industry 4.0 and the Internet of Things . A typical feature of such systems is that they can process large amounts of data in a short time and in are embedded in a higher-level system (“system of systems”). Until 2020, tens of billions of euros have been invested in this technology worldwide.

Cognitive systems should contribute to the added value in the digital transformation of economic processes. They should act as an amplifier of human creativity in innovation processes and support companies in improving products and services.

For data science , cognitive systems are an essential asset for the development of methods and automatic hypotheses . They are used in this context, for example, when designing digital twins or when forecasting events.

conditions

Cognitive systems should be operationally safe and reliable . It must be prevented as much as possible that these systems can be influenced or disabled, for example by cyber attacks . The human-machine interfaces of the systems should be usable and people must not be endangered.

Since a great deal of data is generated during data processing by cognitive systems, which may be stored temporarily or even permanently, the analysis of the data often allows unexpected conclusions to be drawn. Therefore, data protection aspects may have to be taken into account. Users must be informed accordingly about the use of the data by third parties and must then have control over the use of this data.

Demarcation

Cognitive systems can use artificial intelligence processes , but are not to be equated with them.

The generic term for all IT infrastructures , technologies , software solutions and algorithms that make up cognitive systems is cognitive computing . Thus cognitive services can (English: cognitive services ) are provided.

Web links

Individual evidence

  1. ↑ a b c d e Cognitive Machines - Milestone in Knowledge Work - Guide , BITKOM , 2015, accessed on March 6, 2020
  2. ↑ a b c d Cognitive Systems . Fraunhofer Institute for Cognitive Systems IKS. Retrieved March 6, 2020
  3. ↑ Adrian Vogler: The digitization of skill management . In: Computer Science Spectrum . tape 41 , no. 2 , April 1, 2018, p. 97-104 , doi : 10.1007 / s00287-018-1097-y .
  4. ↑ Cognitive assistance systems . German Research Center for Artificial Intelligence. Retrieved March 6, 2020.
  5. ↑ Data - raw material for smart innovations , section "The challenge of data security", Fraunhofer Annual Report 2016, Fraunhofer-Gesellschaft . Retrieved March 9, 2020.