Mineral exploration is getting harder as much of the world has already been surveyed and fewer deposits are economically recoverable. In mineral exploration, there is only a 5% success rate in brownfield exploration, and a smaller 0.5% success rate in greenfield exploration. This is compared to oil & gas exploration, which makes greater use of computer modeling and where there is an 88% success rate.
While a significant amount of capital has been spent on collecting exploration data over the years, it has led to databases that are too vast and complex for geologists to effectively evaluate. At the same time there is not enough data characterizing existing mineral deposits for machine learning techniques to be broadly successful. Mineral deposits are rare and complex with many different attributes.
Minerva believes that the use of AI will change how companies explore for mineral deposits in that they will focus on collecting standardized exploration data in a form that lends itself to interoperability, machine learning (for tasks of perception) and machine reasoning (for tasks requiring cognition). Historical exploration work placed little emphasis on standardization.
Minerva addresses both discovery difficulty and data volume problems, as well as certain metallurgical challenges. Minerva optimizes mineral deposit discovery and certain metallurgical research by combining proprietary and open source AI technologies with other modern technologies such as augmented reality, and with proven legacy technologies such as geological modelling. This optimum combination of AI technologies allows Minerva to find the best locations at which to conduct exploration, to explain why each location was identified, and to provide advice on what additional exploration information to look for at such locations. Minerva is currently engaged in a number of exploration targeting projects incorporating multiple sources of data in both two and three dimensions.
Minerva has carried out a number of projects for government agencies focused on generating public domain exploration targets to promote mining within their jurisdictions. Minerva recently updated a project from 2004 carried out in Canada's Yukon territory. Analysis of the exploration areas highlighted by that project show a very good correlation with claims held for exploration today, fifteen years after the study. Minerva recently launched a new website updating this 2004 study and using its technology to identify exploration targets in the Yukon Territory – www.yukonmineraltargets.com.
Early in 2019, Minerva intends to launch LEO (Language Extender and Organizer), an AI semantics platform customized for minerals exploration companies. This platform will address the need for interoperability between exploration and metallurgical data, which is achievable only through implementation of semantic standards, the latter being a field within which Minerva has specialist expertise. The LEO platform will allow companies to use semantics-based AI systems, such as those provided by Minerva, to greater effect.