Statistical criterion for comparison of binary classifier accuracy

dc.contributor.author	Strebkov E.
dc.contributor.author	Zheltukhin V.
dc.contributor.author	Borodaev I.
dc.date.accessioned	2020-01-15T21:49:01Z
dc.date.available	2020-01-15T21:49:01Z
dc.date.issued	2019
dc.identifier.issn	1742-6588
dc.identifier.uri	https://dspace.kpfu.ru/xmlui/handle/net/156175
dc.description.abstract	© Published under licence by IOP Publishing Ltd. An ABC-method (Accuracy Binary Classifier) for a more accurate assessment of the binary classifier's quality compared to the ROC-method (Receiver Operating Characteristic) is proposed. The ABC-method is suitable for quantitative and qualitative measures in independent and dependent small samples and it is not limited by the laws of distribution of objects in a sample. The ABC-method is effective for classification issues in different scientific and applied fields: IT, physical, technical, medical.
dc.relation.ispartofseries	Journal of Physics: Conference Series
dc.title	Statistical criterion for comparison of binary classifier accuracy
dc.type	Conference Paper
dc.relation.ispartofseries-issue	4
dc.relation.ispartofseries-volume	1158
dc.collection	Публикации сотрудников КФУ
dc.source.id	SCOPUS17426588-2019-1158-4-SID85063810858

Файлы в этом документе

Размер: 46.24Kb

Формат: PDF

Публикации сотрудников КФУ Scopus [24551]
Коллекция содержит публикации сотрудников Казанского федерального (до 2010 года Казанского государственного) университета, проиндексированные в БД Scopus, начиная с 1970г.