GoldSpot Discoveries Successfully Uses Deep Learning to Generate Accurate Geological Maps

Toronto, Ontario--(Newsfile Corp. - October 15, 2019) - GoldSpot Discoveries Corp. (TSXV: SPOT) (the "Company" or "GoldSpot") is pleased to provide an update on the benefits of its proprietary artificial intelligence (AI) technology.

Testimonial From Yamana Gold

"GoldSpot and Yamana Gold recently completed a machine learning collaboration in the area surrounding the El Penon mine site using extensive, multidisciplinary, geological, geophysical and geochemical datasets. The study was successful in identifying known mineralized areas in the mine in blind tests and is now playing a significant role in aggressive ongoing exploration efforts. GoldSpot was able to create a predictive lithological map for covered areas that is particularly useful for prioritizing drill targets. The highly collaborative approach demonstrated by the GoldSpot team contributed greatly to the quality of the final product." Henry Marsden, Senior Vice President, Exploration

Geological maps are a key component of mineral deposit exploration and discovery. These maps are used by geologists to interpret regions around ore deposits, and are generally created by experts using field observations and interpretations. GoldSpot's Research and Development Team has developed proprietary AI techniques, which allow the rapid creation of geological maps for deployment into exploration projects. Combining geology and data science, this application is just one of the use cases for AI in mineral exploration.

Traditionally, geologists create lithological (rock-type) maps by walking the ground and making sense of the composition and structure of the Earth's surface. Always a challenge, this exercise is further complicated when overburden, such as soil or glacial till, overlies the area of interest and potential ore bodies.

To overcome this problem, various methods to perform predictive lithological mapping using remote sensing and geophysical data have been studied and demonstrated. The simplest approach has been for geologists to simply observe datasets which are printed out and layered on top of one another. This kind of manual pattern recognition may be effective when the geologist has experience in a particular area, but it is also a time-consuming and subjective approach which varies depending on the geologist.

GoldSpot has developed Deep Learning AI models for the creation of lithological maps. The technique has shown to be particularly effective in areas where rock outcrop exposure is poor and ore deposit exploration is therefore difficult. This time-saving approach is currently being applied to GoldSpot's clients' projects that include greenfield exploration, as well as brownfields districts where conventional approaches have proven ineffective. The quality and distribution of input data is critical, and can include drone/satellite remote sensing and geophysical data. More specifically, the most successful projects have integrated a selection of radiometrics, magnetics, satellite hyperspectral, synthetic aperture radar, and thermal infrared data.

The results of GoldSpot's predictive geological mapping are typically delivered to clients as georeferenced maps and ESRI shapefiles, which delineate both geological features and a probability map across the area of interest. That probabilistic interpretation offers a significant advantage over traditional maps, which do not quantify the uncertainty of the map-maker's interpretations.

Figure 1: Incomplete Geological Map

To view an enhanced version of Figure 1, please visit:
https://orders.newsfilecorp.com/files/5844/48728_db1ded1164f55a80_001full.jpg

Figure 2: Machine Learning Produced Geological Map

To view an enhanced version of Figure 2, please visit:
https://orders.newsfilecorp.com/files/5844/48728_db1ded1164f55a80_002full.jpg

The use of machine learning to generate accurate lithological maps provides geologists with actionable intelligence, and delivers significant cost and time savings to exploration projects. Maps can be generated early and quickly during the exploration process, and groundtruthed in the field. As field validation continues, new observations are integrated into the AI model. The result is an iterative, machine-assisted feedback loop that allows explorers to move faster than when using slower, manual methods of mapping-based discovery.

Machine learning geological mapping is a single example of GoldSpot's suite of exploration offerings. As a tool in the explorer's toolbox, AI and machine learning are only beginning to be properly recognized as a means of optimizing the entire mineral exploration process.

About GoldSpot Discoveries Corp.

GoldSpot is a technology company that leverages machine learning to reduce capital risk, while working to increase efficiency and success rates in resource exploration and investment. GoldSpot combines proprietary technology with traditional domain expertise, offering a front to-back service solution to its partners. GoldSpot's solutions target big data problems, making full use of historically unutilized data to better comprehend resource property potential. GoldSpot has developed a monetization strategy into multiple verticals of the mining and investment industry, including service offerings, staking and royalty acquisition, and the development of its own artificial-intelligence driven trading platform.

For further information please contact:

Denis Laviolette, President, CEO and Director
GoldSpot Discoveries Corp.
647-992-9837

Neither the TSX Venture Exchange ("TSXV") nor its Regulation Services Provider (as that term is defined in the policies of the TSXV) accepts responsibility for the adequacy or accuracy of this release.

To view the source version of this press release, please visit https://www.newsfilecorp.com/release/48728