Tech moves fast! Stay ahead of the curve with Techopedia!
Join nearly 200,000 subscribers who receive actionable tech insights from Techopedia.
The Mycin project was a computer program used to diagnose infections and figure out which types of bacteria were causing them in patients. The system was actually not used in clinical practice, but it constitutes an excellent early example of a digital expert system and a precursor to much more sophisticated machine learning and knowledge base systems years later.
Mycin was developed at Stanford University in the 1970s. The knowledge base of the Mycin program had around 600 rules. Users would enter answers to a series of “yes” or “no” questions and short answer questions, and the program would eventually choose a weighted probability for a diagnosis. Part of the limitation of this early program was simply computing power – because the program was estimated to take up to half an hour to get through in a clinical environment, it was not considered effective enough to replace human diagnosis at the time. Ethical questions also contributed to the decision not to use Mycin for clinical diagnosis.
However, Mycin has proven to be a stepping stone to more modern systems and described in a book on rule-based expert systems by B. G. Buchanan and E. H. Shortliffe as “the granddaddy of them all” in terms of early artificial intelligence for machine learning systems.