Improving performance of spatio-temporal machine learning models using forward feature selection and target-oriented validation

By Hanna Meyer, Christoph Reudenbach, Tomislav Hengl, M. Katurji, Thomas Nauss. Environmental Modelling & Software, 2018

446 citations recorded by the source corpus.

Subfields: Atmospheric Science, Environmental Engineering, Water Science and Technology

View Paper: https://www.semanticscholar.org/paper/049a599cb522e02eced00946474cdcd1ce66e2fa

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atmospheric Science

In Environmental Engineering

In Water Science and Technology

Citations in the corpus

Performance evaluation and hyperparameter tuning of statistical and machine-learning models using spatial data (2018, methodology): “…importance of bias-reduced spatial resampling methods for performance estimation has been emphasized repeatedly in recent years (Gei et al., 2017; Meyer et al., 2018; Wenger & Olden, 2012), such techniques have not been adopted in all cases (Bui et al., 2015; Pourghasemi & Rahmati, 2018;…”
Predicting into unknown space? Estimating the area of applicability of spatial prediction models (2020, methodology): “We therefore suggested that, in line with the appropriate cross-validation strategy being used ( e.g. spatial, see Meyer et al., 2018; Roberts et al., 2017; Valavi et al., 2018; Pohjankukka et al., 2017; Schratz et al., 2019), the threshold for the AOA is derived from the training data, by…”
Machine learning for digital soil mapping: Applications, challenges and suggested solutions (2020, methodology): “Similarly to the hyperparameter tuning using nested spatial CV, Meyer et al. (2018) showed that autocorrelated covariates lead to overfitting and visible artefacts in the predicted map.”
Random Forest Spatial Interpolation (2020, methodology): “In the precipitation case study, the accuracy was assessed using a “target-oriented” cross-validation strategy [45], i.e., by a nested 5-fold leave-location out cross-validation (LLOCV).”
Importance of spatial predictor variable selection in machine learning applications - Moving from data reproduction to spatial prediction (2019, methodology): “See Meyer et al. (2018) for a more detailed description on this FFS.”