China
Africa
Explore chapters and articles related to this topic
Expert systems
Published in Janet Finlay, Alan Dix, An Introduction to Artificial Intelligence, 2020
PROSPECTOR is an expert system to evaluate geological sites for potential mineral deposits, again developed at Stanford in the late 1970s (Duda et al. 1979). Given a set of observations on the site’s attributes (provided by the user), PROSPECTOR provides a list of minerals, along with probabilities of them being present. In 1984 it was instrumental in discovering a molybdenum deposit worth 100 million dollars! – knowledge representation. Rules, semantic network.– reasoning. Predominantly forward chaining (data-driven), with some backward chaining. Bayesian reasoning is used to deal with uncertainty.– heuristics. Depth first search is focused using the probabilities of each hypothesis.– dialogue/explanation. The dialogue uses mixed control. The user volunteers information at the start of the consultation, and PROSPECTOR can request additional information when required. Explanations are generated by tracing back through the rules that have been fired.
An introduction to BEMSs
Published in G.J. Levermore, Building Energy Management Systems, 2013
The earliest expert system (work started on it in 1964) and a significant advance in expert systems was Dendral, which identifies molecular compounds from their mass spectrograms. It performs better than the best human experts and has made some worthwhile contributions [48]. Prospector was developed in 1978 and is designed to help geologists in their search for ore deposits and to investigate the mineral potential of large areas of land. Tests against known sites of exploration revealed a 7% agreement [49].
Transformation of the Australian mining industry and future prospects
Published in Mining Technology, 2020
The application of artificial intelligence (AI) to mineral exploration is longer than its history would suggest. To the best of our knowledge, the oldest example is the ‘prospector,’ a knowledge-based expert system for mineral exploration developed in the United States in 1979. The system has contributed greatly to the discovery of a molybdenum ore body in Washington, U.S. (Hart et al. 1978; Campbell et al. 1982). Since then, AI-applied mineral exploration technologies have been developed by several researchers using GIS and various soft computing technologies. Recently, there have been a growing number of mineral exploration service companies using AI in Australia, such as Orefox (2020) and Earth AI (2020). Different AI technologies have been applied to mineral exploration depending on the type and nature of data available. In the case of Earth AI, initially, specific exploration models for the target minerals are built based on present primary exploration data in a brownfield. Those trained exploration models can be applied to brownfields that are not included in the artificial intelligence learning process for more detailed exploration or green fields where mineral deposits are likely to be buried. Then, more direct exploration can be carried out based on the results.