GOLDFINDER is a prototype knowledge-based system which assesses the potential of a gold prospect and gives advice on the best locality to site a drill hole. By interaction with the user, the system establishes a dynamic database containing, at a sophisticated level, information about the main lithologies, the nature of any mineralization or alteration, and the structural features of the prospect. Using a method of intersecting loci, the system locates a drill site with a precision governed only by the scale of the prospect map.
GOLDFINDER is written in PROLOG within a Macintosh environment and utilizes an interpreter based on two-valued logic. Knowledge is coded partly as semantic networks but mainly as a production system with a structured rule base. A conventional ‘how’ explanation is improved by appending the reason why the expert included a particular rule. A powerful ‘what if’ facility enables the geologist to change the parameters and use the system for modelling purposes. At present, the domain of GOLDFINDER is restricted to Archaean greenstone belts. The present knowledge base could be removed to provide a shell for prospecting any metalliferous mineral deposit.