Read the latest article “Chemometric recognition of genetically distinct oil families in the Los Angeles basin, California” published in AAPG Bulletin, v. 100, no. 1 (January 2016), pp. 115–135, under authors K. E. Peters, T. L.Wright, L. S. Ramos, J. E. Zumberge, and L. B. Magoon. The article discusses the chemometric analysis and identification of six genetically distinct Miocene tribes (12 families) in the Los Angeles basin. The resulting families were used as a training set to construct a chemometric decision tree used for classification on any additional samples of crude oil or source-rock extract that become available.
There are many instances where a single chemometric model may not provide optimal results for a particular application. In these situations, it may be possible to develop a set of models, each more focused on a particular subset of data. How might these models be deployed? Continue reading