Automated and Emerging Technologies : Artificial Intelligence

How expert systems work

The system asks the user questions to determine the solution / answer. Depending on each answer, the system will ask a different question (e.g. webMD)

Collection of data

A key feature of AI where a program receives data via a user (e.g. speech recognition) or from its sensors (e.g. robot that uses infra-red sensors)

Ability to reason

A key feature of AI where facts are formed from logic and rules. (e.g. All dogs eat meat, Fred is a dog, therefore Fred must eat meat.)

A set of programmed rules

A key feature of AI where stored conditions are used to make decisions. For example, when to slow a self-driving car down.

Expert System

A special type of system that attempts to emulate the expertise of a human (e.g. engineers or doctors).

Machine Learning (supervised)

A user is telling the program what its data means. (e.g. image recognition in Captchas)

Ability to learn and adapt

Although not every AI system will do this (e.g. expert systems), this key feature of AI is necessary for Machine Learning. Systems may be programmed to learn and change their own rules and logic.

User Interface

How questions and statements are output to the user and allows the user to input data.

Machine Learning (unsupervised)

Data is input and the program is learning from the data without the direct aid of a user. The data is plotted on a graph, and then the program identifies which items of data are close to each other.

Inference engine

part of an expert system that makes decisions

Knowledge base

part of an expert system that stores facts

Rule base

part of an expert system that stores the rules based upon the knowledge

artificial intelligence

the development of programs to simulate human behaviour.

Gaming AI

used in computer games to move elements or characters independently based upon the environment

Image recognition

used to identify objects or people in an image

Speech recognition

used to identify words spoken and store them

Natural language recognition

used to receive a command or instruction that is not in a set format and perform the required response

Diagnosis Systems

used, for example, in medical diagnosis