Murray Campbell · Communications of the ACM 1999 · descriptive technical case study · n=700,000 games

Knowledge discovery in deep blue

Cited 80 times in the scientific literature.

Level 5 - mechanism / opinion, no new human data

Level 5 by design analogy; descriptive technical report and expert case overview without comparative experimental controls.

OpenAlex W1997922645 · doi:10.1145/319382.319396 · record verified 2026-08-29

What was done

The authors describe the system design and knowledge-discovery process used by Deep Blue, focusing on its method of extracting usable knowledge from a database of 700,000 Grandmaster chess games.

What was found

The abstract reports that Deep Blue became the first chess computer to defeat a reigning human world chess champion in a regulation match, supported by extracting knowledge from 700,000 Grandmaster games. No additional quantitative performance metrics are provided in the abstract.

Why it matters

It highlights how large-scale knowledge extraction from expert domain datasets can be applied to complex decision-making systems beyond chess.

Limits

The abstract is truncated and provides no detailed algorithmic descriptions, baseline comparisons, or formal quantitative metrics. As a single-system narrative overview, it lacks experimental evaluation.

Cited by