| CPC G01V 1/226 (2013.01) [G01V 1/288 (2013.01); G01V 1/42 (2013.01); G01V 1/50 (2013.01); G06N 3/084 (2013.01); G01V 2210/1234 (2013.01); G01V 2210/1299 (2013.01); G01V 2210/41 (2013.01); G01V 2210/646 (2013.01); G06N 20/00 (2019.01); G06T 5/60 (2024.01); G06T 7/00 (2013.01); G06T 2207/20081 (2013.01)] | 20 Claims | 

| 
               1. A computer-implemented method for creating training data and for training a learning machine in a computer system, comprising: 
            moving a first optimal microseismic event location in a subsurface formation to a first perturbed microseismic event location in each of a plurality of first images; 
                modifying first shear waves and first compressional waves to provide first modified shear waves and first modified compressional waves in each of the first images based on one or more signal travel times between the first perturbed microseismic event location and a fiber optic cable in the subsurface formation to form a plurality of training images configured to train the learning machine, wherein the first modified shear waves and the first modified compressional waves appear to have arisen from the first perturbed microseismic event location; and 
                training the learning machine with the plurality of training images to create a trained learning machine that can output a second optimal microseismic event location in the subsurface formation based on receiving a second image of a second perturbed microseismic event location, the second image comprising second shear waves and second compressional waves. 
               |