A supervised classification method – Cocktail Determination Key and CoDeK

Investor logo

Warning

This publication doesn't include Faculty of Economics and Administration. It includes Faculty of Science. Official publication website can be found on muni.cz.
Authors

LI Ching-Feng ZELENÝ David

Year of publication 2014
Type Appeared in Conference without Proceedings
MU Faculty or unit

Faculty of Science

Citation
Description To synthesize the ecological knowledge from the vegetation plots, one of the important things is a formalization of vegetation classification. A formalized classification is consistent and repeatable by unequivocal assignment rules. Cocktail method was developed to formalize the expert system into algorithms based on the species group concept. Species group concept is the central idea in vegetation classification based on application of the Braun-Blanquet approach. Thus, the unequivocal assignment rules used in Cocktail method can be practically discussed and improved. In this study, a new cocktail method, Cocktail Determination Key, is introduced. This method can be programmed to be applied automatically by the software, CoDeK, which is an application based on the R program (http://bit.ly/cocktail-determination-key). The data input can be done by files using plain text format, or it can be run in the JUICE-R function by JUICE software. The Cocktail Determination Key contains n Cocktail formulae (Xi). While running this key, each vegetation plot is compared with each formula in a stepwise manner (X1 -> X2 -> X3 -> … -> Xn). If the plot fulfils the conditions defined by formula Xi, it is assigned to the vegetation type defined by formula Xi. Cocktail formula in the Cocktail Determination Key contains five main parts: the rank of the formula, species groups, numbers in front of the species groups, logical operators (AND, OR, WITH, WITHOUT, NOT) connecting the species groups, and the code of vegetation unit assigned by this formula. The advantages of Cocktail Determination Key in comparison to other published supervised methods are: it provides the clear argument and logic, based on the presence or absence of species groups in a specific habitat, used in classifying the vegetation; it can be applied in species rich communities; it can be developed step by step without knowing all the existing vegetation types in advance. However, the disadvantages of Cocktail Determination Key are: to construct all the formulae is time consuming and the over-training must be avoided.
Related projects:

You are running an old browser version. We recommend updating your browser to its latest version.