Scientific Discovery: Computational Explorations of the Creative Processes

by Pat Langley, Herbert A. Simon, Gary L. Bradshaw

Cambridge 1987 1st MIT. 8vo., 357pp., hardcover. Fine in Fine DJ.

357 pages, Hardcover

First published January 1, 1987

Reviews

[Lucille Nguyen · Rating 4 out of 5]

A fascinating attempt to utilize production systems/expert systems to generate inductive discoveries in the sciences, particularly historical ones like Newton's laws and Ohms Law. Additionally, it shows the power of heuristic methods to cluster concepts. However, as many reviewers noted at the time (among them Malcolm Forester, Joseph Sneed, and Wolfgang Balzer) this capacity is rather ungeneralizable to the cases of messy data. Nonetheless, the extension of the formal symbol system hypothesis to scientific discovery is itself rather fascinating.

[Gary Robert Gress · Rating 2 out of 5]

An early (1987) AI attempt at typically too high a level. In terms of true contribution and progress, it's probably not much better than curve fitting.