An in-depth analysis of logarithmic data transformation and per-class normalization in machine learning: Application to unsupervised classification of a turbidite system in the Canterbury Basin, New Zealand, and supervised classification of salt in the Eugene Island minibasin, Gulf of Mexico

Abstract

In a machine learning workflow, data normalization is a crucial step that compensates for the large variation in data ranges and averages associated with different types of input measured with different units. However, most machine learning implementations do not provide data normalization beyond the z-score algorithm which subtracts the mean from the distribution and then scales the result by dividing by the standard deviation. Although z-score converts data with Gaussian behavior to have the same shape and size, many of our seismic attribute volumes exhibit log-normal, or even more complicated distributions. Because many machine learning applications are based on Gaussian statistics, we wish to evaluate the impact of more sophisticated data normalization techniques on the resulting classification. To do so, we provide an in-depth analysis of data normalization in machine-learning classifications by formulating and applying a logarithmic data transformation scheme to the unsupervised classifications (including PCA, ICA, SOM, and GTM) of a turbidite channel system in the Canterbury Basin, New Zealand, as well as implementing a per-class normalization scheme to the supervised probabilistic neural network (PNN) classification of salt in the Eugene Island mini-basin, Gulf of Mexico. Compared to the simple z-score normalization, a single logarithmic transformation applied to each input attribute significantly increases the spread of the resulting clusters (and corresponding color contrast), thereby enhancing subtle details in projection and unsupervised classification. However, this same uniform transformation produces less-confident results in supervised classification using probabilistic neural networks. We find that more accurate supervised classifications can be found by applying class-dependent normalization for each input attribute.

Interrelación de mapas geológicos del SGN con inventarios de pozos para uso en la prospección de nuevas perforaciones

“…En la etapa de selección de atributos, se llevó a cabo un análisis exhaustivo para identificar los atributos más relevantes que influyen en la predicción de la productividad de los pozos. Para ello, se utilizó una matriz de correlación para evaluar la relación entre diferentes variables y detectar aquellas con alta colinealidad, es decir, aquellas que están altamente correlacionadas entre sí (Ha et al, 2021).…”

“…Esta transformación fue realizada con el objetivo de reducir la variabilidad de los datos y mejorar la normalidad de la distribución, lo que facilita la interpretación de los resultados y mejora el desempeño de ciertos modelos estadísticos (Ha et al, 2021).…”

Utility of Cognitive Neural Features for Predicting Mental Health Behaviors

“…Finally, we computed the logarithm of the whole dataset obtained so far, which can enhance subtle details in the classification step [58]. Since the domain of logarithms contains only positive real numbers, we calibrated the value of each feature before computing its logarithm.…”