Knowledge (artificial intelligence)

Knowledge in artificial intelligence is information about entities, facts, and relationships that a system can represent and use for reasoning, inference, or action.

Connect

Knowledge representation: Methods for encoding information so that an artificial intelligence system can use it for reasoning or computation. This is the broader discipline concerned with representing AI knowledge.
Automated reasoning: The use of computational procedures to derive conclusions from formalized premises. It is the process by which AI systems draw conclusions from represented knowledge.
Expert system: An AI program that uses a knowledge base and inference rules to solve problems in a specialized domain. It makes explicit domain knowledge central to problem solving.
Symbolic artificial intelligence: An AI approach that represents problems and knowledge with explicit symbols and rules. It is the clearest approach built around manipulable, explicit knowledge.
Knowledge acquisition: The process of gathering, extracting, or encoding information for use in a knowledge-based system. Building useful knowledge bases requires turning scattered information into formal representations.
Propositional logic: A formal system that represents propositions and combines them using logical connectives. It encodes facts as statements that can support deductive reasoning.
Inference engine: A software component that applies rules or logical procedures to derive conclusions from stored information. It applies a knowledge base's rules to produce new facts or answers.
Question answering: The task of producing answers to questions from text, structured data, or other information sources. Knowledge representations let systems connect a question to stored facts and relations.
Machine learning: Methods that enable computer systems to improve performance by learning patterns from data. It often stores learned regularities in model parameters rather than explicit facts.
Knowledge representation and reasoning: The study of representing information in computational forms and deriving conclusions from it. It frames the central challenge of making stored knowledge both expressive and usable.