- Home
- Companies
- Geophysical Insights
- Software
- Paradise - Reveal Stratigraphy Software ...
Paradise - Reveal Stratigraphy Software through Machine Learning
Using the Paradise® AI workbench, geoscientists can generate and analyze seismic data at the sample level, well beyond a wavelet. This powerful capability, along with the application of machine learning to Multi-Attribute Classification, produces profound, sometimes surprising results, particularly the ability to detect features below conventional seismic tuning thickness. Attribute Selection is done through the use of Principal Component Analysis (PCA), and Multi-Attribute Classification is based on the highly robust Self-Organizing Map (SOM) technique, which is applicable with or without well control. This means the workbench tools described are applicable in both exploration and production.

- Highlight reservoirs and stratigraphic features at the sample level
- Detect thin beds and direct hydrocarbon indicators (DHIs)
- Compare machine learning results to well logs to refine results
Pictured Above: Resolution comparison between conventional seismic display and a Paradise multi-attribute SOM classification. (left) Seismic amplitude profile through the 6 well and (right) the SOM results of the same profile identifying the Eagle Ford group that comprises 26 sample-based neuron clusters, which are calibrated to facies and systems tracts. The 2D color map displays the associated neuron cluster colors. Seismic data owned by and provided courtesy of Seitel Inc.
