Model Selection for Machine Learning Algorithm on Decision Making in Oil and Gas Upstream Project Malaysia

Abstract

Model selection is a crucial element in data analysis to get reliable and reproducible statistical inferences or predictions. It is a long history of model selection method arising from research in statistics, information theory, and signal processing. The purpose of this study is to address the problems related to big data in contributing the strategies to make decisions on new investments for upstream Oil and Gas projects in Malaysia. It also discusses the use of machine learning methods for big data processing and highlights current scenarios in a model selection perspective. Machine learning algorithms have proven to work well for statistics used to make decisions. The selection of the machine learning algorithm model does not make drastic assumptions about data, and it can help optimise the exploration process and allow the computer to analyse large amounts of data quickly and accurately. The results show that k-fold cross-validation of the developed model options intended to make subsequent decisions because it is an integral portion of big data processing to gather unexpected new insights, discover new knowledge and improve efficiency.