Lesson 4 of 81972–19926 min
Expert systems and the second winter
In the early 1970s at Stanford, Edward Shortliffe, then an MD and PhD student, built a program called MYCIN to help diagnose serious blood infections and recommend antibiotics. It held its medical knowledge as about 600 separate IF-THEN rules. Tested on real cases, it performed as well as or better than members of Stanford's medical school.
MYCIN was never used on patients, for ethical and legal reasons. But its design escaped the lab. In 1980 Digital Equipment Corporation began using a rule-based system, R1, later called XCON, to configure orders for its VAX-11 computers. AI had found a paying job.
The rest of this lesson is part of Pro
Create a free account to read the first lesson of every path, and go Pro for the rest.
Sources
- Expert systems: MYCIN, EMYCIN and XCON (CPSC 458 lecture notes) — Yale University, Department of Computer Science
- R1 Revisited: Four Years in the Trenches (J. Bachant, J. McDermott, AI Magazine, 1984) — AITopics (AAAI)
- Fifth Generation Computer Project — IPSJ Computer Museum (Information Processing Society of Japan)