TY - JOUR
T1 - Constructing bi-plots for random forest
T2 - Tutorial
AU - Blanchet, Lionel
AU - Vitale, Raffaele
AU - van Vorstenbosch, Robert
AU - Stavropoulos, George
AU - Pender, John
AU - Jonkers, Daisy
AU - van Schooten, Frederik-Jan
AU - Smolinska, Agnieszka
N1 - Funding Information:
A/Prof John Penders is an expert in molecular epidemiology and microbial ecology. His research group integrates metagenomic methods within the context of prospective epidemiological studies using various longitudinal statistical and bioinformatics tools to elucidate the role of the microbiome in health and disease. His group (at Maastricht University, The Netherlands) is currently funded by The Netherlands Organization for Scientific Research, The Netherlands Organization for Health Research and Development and the Joint Programming Initiative on Healthy Diet for Healthy Living. He has authored more than 120 publications, including in leading journals like Nature Biotechnology, Lancet Infectious Diseases, Gastroenterology, Gut, Mucosal Immunology and Microbiome.
Funding Information:
This work was supported by Netherlands Organisation for Scientific Research (NWO, the Netherlands) (grant number: 016.Veni.178.064 ).
Publisher Copyright:
© 2020 Elsevier B.V.
PY - 2020/9/22
Y1 - 2020/9/22
N2 - Current technological developments have allowed for a significant increase and availability of data. Consequently, this has opened enormous opportunities for the machine learning and data science field, translating into the development of new algorithms in a wide range of applications in medical, biomedical, daily-life, and national security areas. Ensemble techniques are among the pillars of the machine learning field, and they can be defined as approaches in which multiple, complex, independent/uncorrelated, predictive models are subsequently combined by either averaging or voting to yield a higher model performance. Random forest (RF), a popular ensemble method, has been successfully applied in various domains due to its ability to build predictive models with high certainty and little necessity of model optimization. RF provides both a predictive model and an estimation of the variable importance. However, the estimation of the variable importance is based on thousands of trees, and therefore, it does not specify which variable is important for which sample group.The present study demonstrates an approach based on the pseudo-sample principle that allows for construction of bi-plots (i.e. spin plots) associated with RF models. The pseudo-sample principle for RF. is explained and demonstrated by using two simulated datasets, and three different types of real data, which include political sciences, food chemistry and the human microbiome data. The pseudo-sample bi plots, associated with RF and its unsupervised version, allow for a versatile visualization of multivariate models, and the variable importance and the relation among them. (c) 2020 Elsevier B.V. All rights reserved.
AB - Current technological developments have allowed for a significant increase and availability of data. Consequently, this has opened enormous opportunities for the machine learning and data science field, translating into the development of new algorithms in a wide range of applications in medical, biomedical, daily-life, and national security areas. Ensemble techniques are among the pillars of the machine learning field, and they can be defined as approaches in which multiple, complex, independent/uncorrelated, predictive models are subsequently combined by either averaging or voting to yield a higher model performance. Random forest (RF), a popular ensemble method, has been successfully applied in various domains due to its ability to build predictive models with high certainty and little necessity of model optimization. RF provides both a predictive model and an estimation of the variable importance. However, the estimation of the variable importance is based on thousands of trees, and therefore, it does not specify which variable is important for which sample group.The present study demonstrates an approach based on the pseudo-sample principle that allows for construction of bi-plots (i.e. spin plots) associated with RF models. The pseudo-sample principle for RF. is explained and demonstrated by using two simulated datasets, and three different types of real data, which include political sciences, food chemistry and the human microbiome data. The pseudo-sample bi plots, associated with RF and its unsupervised version, allow for a versatile visualization of multivariate models, and the variable importance and the relation among them. (c) 2020 Elsevier B.V. All rights reserved.
KW - Random forest interpretation
KW - Pseudo samples
KW - Bi-plots
KW - Proximity matrix
KW - Principal coordinates analysis
KW - PSEUDO-SAMPLE TRAJECTORIES
KW - PARTIAL LEAST-SQUARES
KW - FAULT-DIAGNOSIS
KW - KERNEL
U2 - 10.1016/j.aca.2020.06.043
DO - 10.1016/j.aca.2020.06.043
M3 - Article
C2 - 32928475
SN - 0003-2670
VL - 1131
SP - 146
EP - 155
JO - Analytica Chimica Acta
JF - Analytica Chimica Acta
ER -