This presentation explores the many categories of seismic attributes created in the last 20 years and their general use in an interpretation workflow.

Unsupervised Neural Analysis of seismic attributes has been shown to be effective in understanding variations in unconventional resource geological deposition, finding "sweet spots" and understanding complex structural and fracture trends. Neurons find natural clusters in the data and classify into Self-Organized Maps. A neural map is a 2D representation of the result of classifying and associating the data, which may be in "n" dimensions, such as many attributes in a 3D volume. A series of case histories, both unconventional and conventional in nature are shown in which neural mapping have helped find production, understand reservoir properties, fracture trends and even pressure zones in data.