DENDRAL was the Stanford expert system that analyzed mass-spectrometry data to identify organic molecules.
Developed during the 1960s, DENDRAL combined chemical knowledge with heuristic search. Its name came from “dendritic algorithm,” reflecting the branching structure of its search process. The project was led by Edward Feigenbaum, Bruce Buchanan, Joshua Lederberg, and collaborators.
DENDRAL is important because it showed that an AI system could perform useful specialist reasoning when supplied with a carefully organized knowledge base. It did not simply memorize chemical answers; it generated and tested possible molecular structures against spectrographic evidence. A common mix-up is confusing DENDRAL with MYCIN, another Stanford expert system that focused on infectious diseases and antibiotic recommendations. DENDRAL’s influence helped establish the expert-system tradition in artificial intelligence and inspired later systems designed for medicine, engineering, and business decision-making.