Logic Programming as a Representation of Knowledge

• Published in: Computer (Long Beach, Calif.) Vol. 16; no. 10; pp. 106 - 111
• Main Author: DAHL, V
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.10.1983; IEEE Computer Society
• Subjects: Applied sciences; Artificial intelligence; Computer languages; Computer science; control theory; systems; Deductive databases; Exact sciences and technology; Expert systems; Functional programming; Learning and adaptive systems; Logic programming; Natural language processing; Robots; Writing; Knowledge representation
• ISSN: 0018-9162
• DOI: 10.1109/MC.1983.1654206

The idea that logic could serve as a programming language was put to practical use around 1972 in the form of Prolog. It has proved extremely valuable in many ambitious and diverse computational areas, including natural language processing, deductive databases, robotics, symbolic integration, and expert systems. Logic has traditionally provided a firm conceptual framework for representing knowledge, as it can formally deal with the notion of logical consequence. The introduction of Prolog has made it possible to represent knowledge in terms of logic and also to expect appropriate inferences to be drawn from it automatically. This article illustrates and explores these ideas with respect to two central representational issues: problem-solving knowledge and database knowledge. The technical aspects of both subject have been covered elsewhere; this explanation uses simple, nontechnical terms.