A taxonomy of Collective Machine Learning Models applicable to well logging predictive anticipation

Abstract

Traditional methods of measuring well logs are expensive, error-prone and time-consuming, which has led to the development of machine learning models that can predict well logging based on well-log data. This study aims to determine the most effective machine learning models for predicting of well logging based on available well-log data. The study covers a detailed explanation of the data-gathering and pre-processing techniques used. Features were used in the models, namely gamma ray (GR), bulk density (RHOB), neutron porosity (NPHI), resistivity (RT), spontaneous potential (SP), trained and evaluated based on their performance, namely linear regression, support vector machine SVM, Neural Network NN and decision Trees DT models. The models were evaluated based on their Mean Squared Error, R squared, Mean Absolute Error and RMSE values. Our results showed that the Decision Trees (DT) for MSE value of 10.86, achieving a Root Mean Squared Error (RMSE) value of 3.29, MAE value of 2.225 and R square value of 0.59. These findings suggest that machine learning models can be a powerful tool for predicting of best training from well-log data, in particular, holds great promise for future modelling efforts in this area.