English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In the literature, a large body of work advocates the use of log-ratiotransformation for multivariate statistical analysis of compositional data. Incontrast, few studies have looked at how data transformation changes theefficacy of machine learning classifiers within geoscience. This letterpresents experiment results and empirical observations to further explore thisissue. The objective is to study the effects of data transformation on geozoneclassification performance when machine learning (ML) classifiers/estimatorsare trained using geochemical data. The training input consists of explorationhole assay samples obtained from a Pilbara iron-ore deposit in WesternAustralia, and geozone labels assigned based on stratigraphic units, theabsence or presence and type of mineralization. The ML techniques consideredare multinomial logistic regression, Gaussian na\"{i}ve Bayes, kNN, linearsupport vector classifier, RBF-SVM, gradient boosting and extreme GB, randomforest (RF) and multi-layer perceptron (MLP). The transformations examinedinclude isometric log-ratio (ILR), center log-ratio (CLR) coupled withprincipal component analysis (PCA) or independent component analysis (ICA), anda manifold learning approach based on local linear embedding (LLE). The resultsreveal that different ML classifiers exhibit varying sensitivity to thesetransformations, with some clearly more advantageous or deleterious thanothers. Overall, the best performing candidate is ILR which is unsurprisingconsidering the compositional nature of the data. The performance of pairwiselog-ratio (PWLR) transformation is better than ILR for ensemble and tree-basedlearners such as boosting and RF; but worse for MLP, SVM and other classifiers.

Empirical observations on the effects of data transformation in machine learning classification of geological domains