Decision Trees in Ecological Modelling - Modelling Complex Ecological Dynamics - page 203

Table 14.1 An overview of decision tree types and systems for learning them, with respect to the
number and type of target variables (targets)
Type of targets (decision trees) Number of targets

Single-target Multi-target

Discrete (classification trees) C4.5, C5.0, J4.8, CART, CLUS CLUS

Continuous (regression trees) CART, CLUS CLUS

M5, M5 0 , Cubist

Continuous (model trees) MT-SMOTI

commonly used implementation of classification trees is likely J4.8, the Java
reimplementation of C4.5 within the WEKA suite (Witten and Frank 2005).
Besides CART, the M5 system (Quinlan 1992) builds regression trees. As
compared to CART, the novelty in M5 is that it can also build model trees (with
linear models in the leaves). The commercial successor of M5 is Cubist (RuleQuest
2009), which transcribes the learned regression and model trees into rules (which
are further post-processed/simplified). The publicly available reimplementation of
M5 is called M5 0 and is part of the WEKA suite (Witten and Frank 2005).
The construction of multi-target trees is implemented in the software system
CLUS (Blockeel and Struyf 2002; Struyf and Dˇeroski 2006; Struyf et al. 2010).
CLUS can build trees predicting a single target or multiple targets. It can also
consider discrete and continuous targets, i.e., can build multi-target classification
and regression trees. The system MT-SMOTI (Appice and Dzeroski 2007) builds
multi-target model trees, whose leaves can contain multiple linear equations for
predicting the values of each target.
An overview of the different systems for building different types of decision
trees is given in Table 14.1 .

14.4 Modelling Population Dynamics with Decision Tree
Approaches

Population dynamics studies changes of the size and structure of populations over
time, taking into account environmental and biological processes influencing these
changes. For example, one might study the size of a brown bear population as
affected by its initial size, sex and age structure, reproduction age, fertility and
mortality of different age classes. The modelling formalism most often used by
ecological experts is the formalism of differential equations, which describe the
change of state of a dynamic system over time (see Chaps. 6, 7, 9). A typical
approach to modelling population dynamics can be as follows: an ecological expert
writes a set of differential equations that capture the most important relationships in
the domain. These are often linear differential equations. The coefficients of these
equations are then determined (calibrated) using measured data.
Relationships among attributes describing internal demographic properties of a
population and the set of external environmental attributes influencing changes of