Summary Fuzzy Modeling

At the center of all fuzzy modeling of ecological problems is located a rule-based system. The inputs for this system are taken from a GIS describing a landscape or from a dynamic simulation system. The outputs of the model evaluate environmental indicators. The rule set is easy to understand and should be a basis for discussing the model with other experts. Use of a fuzzy model should be accompanied by a strategy for its further development. This strategy should encompass visualization of system behavior, test on real data sets, sensitivity analysis, etc., including an adaptation using a training algorithm.

The following advices can be offered on the development of fuzzy models:

• start with simple models (no more than three inputs, otherwise use two models);

• use triangular- or trapezoidal-shaped membership functions;

• use FUZZY_AND Prod instead of FUZZY_AND Min;

• use crisp values as output type;

• use training procedures to enhance the model; and

• check the model using graphical analysis tools, rule statistics, sensitivity analyses.

See also: Agriculture Models; Agriculture Systems; Application of Ecological Informatics; Artificial Neural Networks; Biodiversity; Biological Integrity; Biomass; Computer Languages; Development Capacity; Ecological Network Analysis, Environ Analysis; Ecosystems; Empirical Models; Habitat; Habitat Selection and Habitat Suitability Preferences; Landscape Modeling; Multilayer Perceptron; Sensitivity and Uncertainty; Software; Spatial Distribution Models; Spatial Models and Geographic Information Systems; Systems Ecology.

: gl (p2 ■ p3 ■ ...;?l; ?2 ■ ...;bl; b2 ■ ...;t ) g 2(pl^ p3 ■ ...;?!■ ?2 ■ ...;bu ...;t )

Was this article helpful?

0 0
Project Earth Conservation

Project Earth Conservation

Get All The Support And Guidance You Need To Be A Success At Helping Save The Earth. This Book Is One Of The Most Valuable Resources In The World When It Comes To How To Recycle to Create a Better Future for Our Children.

Get My Free Ebook


Post a comment