Mushroom data creation, curation, and simulation to support classification tasks

Abstract Predicting if a set of mushrooms is edible or not corresponds to the task of classifying them into two groups—edible or poisonous—on the basis of a classification rule. To support this binary task, we have collected the largest and most comprehensive attribute based data available. In this...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Dennis Wagner, Dominik Heider, Georges Hattab
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/a12405e30d7941f99e3a835a1d4731ae
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:a12405e30d7941f99e3a835a1d4731ae
record_format	dspace
spelling	oai:doaj.org-article:a12405e30d7941f99e3a835a1d4731ae2021-12-02T14:27:53ZMushroom data creation, curation, and simulation to support classification tasks10.1038/s41598-021-87602-32045-2322https://doaj.org/article/a12405e30d7941f99e3a835a1d4731ae2021-04-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-87602-3https://doaj.org/toc/2045-2322Abstract Predicting if a set of mushrooms is edible or not corresponds to the task of classifying them into two groups—edible or poisonous—on the basis of a classification rule. To support this binary task, we have collected the largest and most comprehensive attribute based data available. In this work, we detail the creation, curation and simulation of a data set for binary classification. Thanks to natural language processing, the primary data are based on a text book for mushroom identification and contain 173 species from 23 families. While the secondary data comprise simulated or hypothetical entries that are structurally comparable to the 1987 data, it serves as pilot data for classification tasks. We evaluated different machine learning algorithms, namely, naive Bayes, logistic regression, and linear discriminant analysis (LDA), and random forests (RF). We found that the RF provided the best results with a five-fold Cross-Validation accuracy and F2-score of 1.0 ( $$\mu =1$$ μ = 1 , $$\sigma =0$$ σ = 0 ), respectively. The results of our pilot are conclusive and indicate that our data were not linearly separable. Unlike the 1987 data which showed good results using a linear decision boundary with the LDA. Our data set contains 23 families and is the largest available. We further provide a fully reproducible workflow and provide the data under the FAIR principles.Dennis WagnerDominik HeiderGeorges HattabNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-12 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Dennis Wagner Dominik Heider Georges Hattab Mushroom data creation, curation, and simulation to support classification tasks
description	Abstract Predicting if a set of mushrooms is edible or not corresponds to the task of classifying them into two groups—edible or poisonous—on the basis of a classification rule. To support this binary task, we have collected the largest and most comprehensive attribute based data available. In this work, we detail the creation, curation and simulation of a data set for binary classification. Thanks to natural language processing, the primary data are based on a text book for mushroom identification and contain 173 species from 23 families. While the secondary data comprise simulated or hypothetical entries that are structurally comparable to the 1987 data, it serves as pilot data for classification tasks. We evaluated different machine learning algorithms, namely, naive Bayes, logistic regression, and linear discriminant analysis (LDA), and random forests (RF). We found that the RF provided the best results with a five-fold Cross-Validation accuracy and F2-score of 1.0 ( $$\mu =1$$ μ = 1 , $$\sigma =0$$ σ = 0 ), respectively. The results of our pilot are conclusive and indicate that our data were not linearly separable. Unlike the 1987 data which showed good results using a linear decision boundary with the LDA. Our data set contains 23 families and is the largest available. We further provide a fully reproducible workflow and provide the data under the FAIR principles.
format	article
author	Dennis Wagner Dominik Heider Georges Hattab
author_facet	Dennis Wagner Dominik Heider Georges Hattab
author_sort	Dennis Wagner
title	Mushroom data creation, curation, and simulation to support classification tasks
title_short	Mushroom data creation, curation, and simulation to support classification tasks
title_full	Mushroom data creation, curation, and simulation to support classification tasks
title_fullStr	Mushroom data creation, curation, and simulation to support classification tasks
title_full_unstemmed	Mushroom data creation, curation, and simulation to support classification tasks
title_sort	mushroom data creation, curation, and simulation to support classification tasks
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/a12405e30d7941f99e3a835a1d4731ae
work_keys_str_mv	AT denniswagner mushroomdatacreationcurationandsimulationtosupportclassificationtasks AT dominikheider mushroomdatacreationcurationandsimulationtosupportclassificationtasks AT georgeshattab mushroomdatacreationcurationandsimulationtosupportclassificationtasks
_version_	1718391270019170304

Mushroom data creation, curation, and simulation to support classification tasks

Ejemplares similares