Traditionally mapping of soil properties, soil type and soil condition has been done by in-situ augur or core samples. This method is slow, costly, and subjective depending on the experience of the operator and the requirements of the survey. Often these sample locations are randomly selected, or limited to 1 per unit area, from within a site which may not fully represent the spatial variability of the soil property of interest. Furthermore, it is difficult to create maps from point samples as boundaries are rarely sampled and must be inferred. A CMD Mini-Explorer 6L instrument has been used to survey a farm (circa. 32 hectares) in the summer of 2021. This instrument provides 6 layers of apparent conductivity measurements reaching a depth of approx. 2.5m. These six 1m x 1m gridded maps of electrical conductivity act as input to a machine learning and GIS based analysis. In this case, we attempt to link these data layers to a national soil classification scheme using surveyed in-situ augur points where the soil classifications present are well understood. We use modern machine learning methods to identify the pattern of layered electrical conductivity associated with each soil classification present at an auger site. This pattern can then be used to predict the soil class in other areas on the farm resulting in a high-resolution map of soil type across the farm which could be used for precision agriculture applications such as fertilizer management, crop rotation or drainage planning.