The database: a prerequisite for large-scale use of artificial intelligence in mining

Paper presented at a meeting on: Artificial intelligence in the minerals sector, held in Nottingham, UK, 20 April 1993 (original title: The database: a pre-requisite for large-scale use of AI in mining). Artificial intelligence is a term which has been widely applied to different techniques including rule-based systems using inductive and deductive logic algorithms, neural networks, learning systems and genetic algorithms. In principle it should be applied to adaptive statistical techniques. Some of these applications have been coded in languages which are specifically suited to the novel requirements of AI but are not easy to integrate with more traditional computing methods. Efforts have been made to bridge the gaps, particularly with database management systems and such efforts at integration should provide the key to future success in the application of AI. A field which has been termed "knowledge engineering" cannot be credible as such if it fails to provide easy access to raw materials of "knowledge" i.e. the data residing in databases. In the mining industry there are a number of significant areas in which AI could usefully be applied throughout the industry, but its advance is slow because of the lack or inefficiency of such database links. Current development of object-oriented database systems for geology and mining holds out hope for such integration in the foreseeable future