solve the problem

Quantum Geology System hereafter called “QGS” is a system of using artificial intelligence in the forms of machine learning and cognitive analysis for rapidly generating quality exploration targets for minerals, that compares the data of thousands of known deposits against the data of areas to be explored. The system looks for obvious correlations in data, hidden patterns, clusters and relationships, across massive data sets.

So how's it work?
We teach the system, by showing it known gold mines, plus up to 100 layers of geological data about each mine.
The system then finds unique signatures for each mine- it can find up to 16 million- which is a few more than a geologist could find.


layer image64

How do you know it works?

We test the system by asking it to find other known gold mines, that it hasn’t been trained on.

 

 

testing

 

 

 The output of the Quantum Geology System is a prospectivity map and spatial data showing how the target exploration areas compare to known mineral deposits from around the world.  This allows geoscientists to decrease risk in exploration and enhances the efficiency of exploration programs. 

The QGS is made up of a number of algorithms and data sets

  1. Data preparation Algorithm DPA
  2. Training Data Set Algorithm (TDSA)
  3. Mineral Search Algorithm MSA
  4. Signature Score Algorithm SSA)

Example Data Set Layers

  1. Landsat images
  2. ASTER images
  3. Geological maps
  4. Spatial data sets of known minerals and elements and pathfinder elements
  5. Geochemistry
  6. Radiometric data
  7. Drill hole data
  8. Historic mining data
  9. Magnetic data
  10. Topographic data

 The starting point for the QGS is the Data Preparation algorithm (DPA), which takes the raw data layers and prepares them for input to the Training Data Set Algorithm. Initially the DPA creates a number of new data layers using digital imaging processing (DIP)and other proprietary methods.  The DPA also carries out operations such as multivariate analysis and Automatic clustering analysis on the relative data.

THE DPA increases the number of data layers to usually just over 100, depending on the amount of initial raw data layers.

The DPA data set is now fed into the Training Data Set Algorithm (TSDA), that is used to train the Mineral Search Algorithm system.

For the TSDA, a spatial data set is created by inputting a set of spatial data on known mineral deposits, mines and occurrences. 

This spatial set is created by geoscientists and grouped and ranked, with the highest ranking being for large economic mineable deposits through to the lowest rank of simple mineral occurrence.

The DPA data set layers are added to the TSDA, the algorithm then uses machine learning and cognitive analysis to create a neural processing chain which allocates each spatially separate area a mineral geologic model. Validation and testing is then done to test the accuracy of these model files.

TSDA processing is only carried out once per mineral or element that is being sought.

Mineral Search Algorithm (MSA)

The MSA is the largest and most intensive algorithm in the QGS system.

The target area spatial boundaries and target raw data layers are inputted to the Data Preparation Algorithm (DPA) which prepares the raw data to be processed by the MSA.