A generalized linear model, generalized additive model, maximum entropy and boosted regression tree methods were used in the analyses.
In the BioCreative V, several machine learning models have been explored for the CID task, including Naïve Bayes [30], maximum entropy [14, 19], logistic regression [21], and support vector machine (SVM).
The seven different SDM methods used in analyses include: Bioclim, Domain, generalized linear model (GLM), generalized additive model (GAM), maximum entropy model (Maxent), boosted regression tree (BRT), and random forests (RF).
Logistic regression and maximum entropy (MaxEnt) models were assessed independently and evaluated as predictive tools to test the value of presence/absence and presence only data in predicting species distributions.
Using generalized kernel ridge regression and maximum entropy models, others [ 17, 18] have described a general and interesting concept, which is the fitness landscape.
We report a comparison of eight representative methods, multifactor dimensionality reduction (MDR), full interaction model (FIM), information gain (IG), Bayesian epistasis association mapping (BEAM), SNP harvester (SH), maximum entropy conditional probability modeling (MECPM), logistic regression with an interaction term (LRIM), and logistic regression (LR).
The first, with mathematical roots dating to the 1950s, is maximum entropy (MaxEnt), also called (multinomial) logistic regression (e.g., Ratnaparkhi 1997).
Examples of correlative habitat suitability models include BRT (Boosted Regression Tree), MaxEnt (Maximum Entropy), and CART (Classification and Regression Tree) models, which rely on occurrence data.
We analyzed simulated data using discrete choice, logistic-regression, and a maximum entropy method (Maxent).
To identify the best method, model performance and distributional accuracy was assessed and compared using three approaches: Generalised Additive Models (GAM), Boosted Regression Trees (BRT) and Maximum Entropy (MaxEnt).
