Supervised classification of plant communities with artificial
neural networks

J 2005

Supervised classification of plant communities with artificial neural networks

ČERNÁ, Lenka a Milan CHYTRÝ

Základní údaje

Originální název

Supervised classification of plant communities with artificial neural networks

Název česky

Řízená klasifikace rostlinných společenstev pomocí umělých neuronových sítí

Autoři

ČERNÁ, Lenka (203 Česká republika) a Milan CHYTRÝ (203 Česká republika, garant)

Vydání

Journal of Vegetation Science, Uppsala, Opulus Press, 2005, 110-9233

Další údaje

Jazyk

angličtina

Typ výsledku

Článek v odborném periodiku

Obor

10600 1.6 Biological sciences

Stát vydavatele

Česká republika

Utajení

není předmětem státního či obchodního tajemství

Odkazy

URL

Kód RIV

RIV/00216224:14310/05:00012592

Organizační jednotka

Přírodovědecká fakulta

UT WoS

000232457000006

Klíčová slova anglicky

Cluster analysis; Grassland; Multi-layer perceptron; Phytosociological data; Predictive habitat modelling; Vegetation survey

Štítky

cluster analysis, grassland, Multi-layer perceptron, Phytosociological data, Predictive habitat modelling, Vegetation survey

Změněno: 16. 12. 2006 18:44, prof. RNDr. Milan Chytrý, Ph.D.

Anotace

ORIG CZ

V originále

Questions: Are artificial neural networks useful for the automatic assignment of species composition records from vegetation plots to a priori established classes (vegetation units)? Is the assignment more accurate (1) if the classes are defined by numerical classification rather than by expert-based classification; (2) if the training data set is selected to include plots that are richer in diagnostic species of particular classes? Material: Species composition records (relevés) from 4186 plots of Czech grasslands. Methods: Plots were classified into 11 phytosociological alliances (expert classification) and into 11 clusters derived from numerical cluster analysis. Some plots were used for training the classifiers, which were the multi-layer perceptrons (MLP; a type of artificial neural network). Other plots were used for testing the performance of these classifiers. Plots used for training were selected (1) randomly; (2) according to higher representation of diagnostic species of particular classes. Results: Different MLP classifiers correctly classified 77-83% of plots to the classes of the expert classification and 70-78% to the classes of the numerical classification. The better result in the former case was mainly due to two classes in the expert classification, which were well recognized by the classifiers and at the same time contained a large proportion of the plots of the entire data set. Correct classification of the plots belonging to these large classes resulted in a good overall performance of the classifiers. After training with randomly chosen plots, the classifiers produced better results than after training with plots that contained more diagnostic species. This indicates that the biased selection of the training plots disables the classifiers to recognize the entire variation within the classes and results in errors when new plots are to be classified. Conclusions: MLP is suitable for assigning vegetation plots to already established classes. Unlike some other methods of supervised classification, it performs well even in communities that are poor in diagnostic species. However, the method does not provide clear assignment keys that could be used for class identification in field surveys. It is therefore more appropriate in applications that aim at a reliable class assignment rather than understanding the assignment rules.

Česky

Testování umělých neuronových sítí jako metody řízená klasifikace rostlinných společenstev

Návaznosti

GA206/02/0957, projekt VaV

Název: Formalizovaná klasifikace polopřirozené travinné vegetace České republiky

Investor: Grantová agentura ČR, Formalizovaná klasifikace polopřirozené travinné vegetace České republiky

MSM0021622416, záměr

Název: Diverzita biotických společenstev a populací: kauzální analýza variability v prostoru a čase

Investor: Ministerstvo školství, mládeže a tělovýchovy ČR, Diverzita biotických společenstev: kauzální analýza variability v prostoru a čase

Citovat

ČERNÁ, Lenka a Milan CHYTRÝ. Supervised classification of plant communities with artificial neural networks. Journal of Vegetation Science. Uppsala: Opulus Press, 2005, roč. 16, č. 4, s. 407-414. ISSN 110-9233.

@article{575062,
   author = {Černá, Lenka and Chytrý, Milan},
   article_location = {Uppsala},
   article_number = {4},
   keywords = {Cluster analysis; Grassland; Multi-layer perceptron; Phytosociological data; Predictive habitat modelling; Vegetation survey},
   language = {eng},
   issn = {110-9233},
   journal = {Journal of Vegetation Science},
   title = {Supervised classification of plant communities with artificial neural networks},
   url = {http://www.sci.muni.cz/botany/chytry/Cerna-Chytry2005_JVS.pdf},
   volume = {16},
   year = {2005}
}

TY  - JOUR
ID  - 575062
AU  - Černá, Lenka - Chytrý, Milan
PY  - 2005
TI  - Supervised classification of plant communities with artificial neural networks
JF  - Journal of Vegetation Science
VL  - 16
IS  - 4
SP  - 407-414
EP  - 407-414
PB  - Opulus Press
SN  - 1109233
KW  - Cluster analysis
KW  - Grassland
KW  - Multi-layer perceptron
KW  - Phytosociological data
KW  - Predictive habitat modelling
KW  - Vegetation survey
UR  - http://www.sci.muni.cz/botany/chytry/Cerna-Chytry2005_JVS.pdf
N2  - Questions: Are artificial neural networks useful for the automatic assignment of species composition records from vegetation plots to a priori established classes (vegetation units)? Is the assignment more accurate (1) if the classes are defined by numerical classification rather than by expert-based classification; (2) if the training data set is selected to include plots that are richer in diagnostic species of particular classes? Material: Species composition records (relevés) from 4186 plots of Czech grasslands. Methods: Plots were classified into 11 phytosociological alliances (expert classification) and into 11 clusters derived from numerical cluster analysis. Some plots were used for training the classifiers, which were the multi-layer perceptrons (MLP; a type of artificial neural network). Other plots were used for testing the performance of these classifiers. Plots used for training were selected (1) randomly; (2) according to higher representation of diagnostic species of particular classes. Results: Different MLP classifiers correctly classified 77-83% of plots to the classes of the expert classification and 70-78% to the classes of the numerical classification. The better result in the former case was mainly due to two classes in the expert classification, which were well recognized by the classifiers and at the same time contained a large proportion of the plots of the entire data set. Correct classification of the plots belonging to these large classes resulted in a good overall performance of the classifiers. After training with randomly chosen plots, the classifiers produced better results than after training with plots that contained more diagnostic species. This indicates that the biased selection of the training plots disables the classifiers to recognize the entire variation within the classes and results in errors when new plots are to be classified. Conclusions: MLP is suitable for assigning vegetation plots to already established classes. Unlike some other methods of supervised classification, it performs well even in communities that are poor in diagnostic species. However, the method does not provide clear assignment keys that could be used for class identification in field surveys. It is therefore more appropriate in applications that aim at a reliable class assignment rather than understanding the assignment rules.
ER  -

ČERNÁ, Lenka a Milan CHYTRÝ. Supervised classification of plant communities with artificial neural networks. \textit{Journal of Vegetation Science}. Uppsala: Opulus Press, 2005, roč.~16, č.~4, s.~407-414. ISSN~110-9233.

Podrobný výpis o publikaci