Comparison and ranking of different modelling techniques for prediction of site index in Mediterranean mountain forests

By Wim Aertsen, V. Kint, Jos Van Orshoven, K. Ozkan, Bart Muys. 2010

407 citations recorded by the source corpus.

Subfields: Environmental Engineering, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/ca245e5a65f383fae9ea92a92a31b00199ccc8cd

How researchers see this paper

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

In Environmental Engineering

In Nature and Landscape Conservation

Citations in the corpus

Performance evaluation and hyperparameter tuning of statistical and machine-learning models using spatial data (2018, background): “…an ongoing discussion about the usage of parametric vs. non-parametric models in the ﬁeld of ecological modeling (Perretti & Munch, 2015), most studies prefer parametric ones due to the ability to interpret relationships between the predictors and the response (Aertsen et al., 2010; Jabot, 2015).”
Forest Growth and Yield Modeling (2011, methodology): “…classification 292 CH 14 MODEL CONSTRUCTION and regression trees (CART; Nigh and Love, 2004; Fan et al., 2006; R€aty and Kangas, 2008), generalized additive models (GAMs; Zhang et al., 2008; Aertsen et al., 2010), and k-nearest neighbor methods (k-NN; Temesgen et al., 2003; Sironen et al., 2008).”
Application and Comparison of Decision Tree-Based Machine Learning Methods in Landside Susceptibility Assessment at Pauri Garhwal Area, Uttarakhand, India (2017, methodology): “…need assumptions on the distribution of explicative factors; (ii) it is capable of calculating interaction between factors; (iii) the random predictor selection used in RF holds low bias; and (iv) it is able to deal with unbalanced data and over-fitting (Aertsen et al. 2010; Prasad et al. 2006).”
GIS-based groundwater potential mapping using boosted regression tree, classification and regression tree, and random forest machine learning models in Iran (2015, methodology): “…26.59 301.00 49.75 1.87 7.32–14.99 439,780 14.20 200.00 33.06 2.33 Fault density (km/km2) <8.52 2,096,826 67.69 382.00 63.14 0.93 8.52–25.57 607,940 19.63 111.00 18.35 0.93 25.57–42.39 363,765 11.74 111.00 18.35 1.56 42.39–58.97 29,082 0.94 1.00 0.17 0.18 among predictors (Aertsen et al. 2010).”
Estimating forest soil bulk density using boosted regression modelling (2010, background): “It could be argued that selecting simple models is important not only for reducing the cost of collecting irrelevant data when using the model for prediction, but also for reducing the risk of overfitting, which increases with model complexity (Aertsen et al., 2010).”