Show simple item record
dc.contributor.author |
Škrabánek, Pavel
|
cze |
dc.contributor.author |
Doležel, Petr
|
cze |
dc.date.accessioned |
2020-03-19T12:22:39Z |
|
dc.date.available |
2020-03-19T12:22:39Z |
|
dc.date.issued |
2017 |
eng |
dc.identifier.issn |
1211-555X |
eng |
dc.identifier.uri |
https://hdl.handle.net/10195/74669 |
|
dc.description.abstract |
In this contribution, the question of reporting performance of binary classifiers is opened in context of the so called class imbalance problem. The class imbalance problem arises when a dataset with a highly imbalanced class distribution is used within the training or evaluation process. In such cases, only measures, which are not biased by distribution of classes in datasets, should be used; however, they cannot be chosen arbitrarily. They should be selected so that their outcomes provide desired information; and simultaneously, they should allow a full comparison of just evaluated classifier performance along, with performances of other solutions. As is shown in this article, the dilemma with reporting performance of binary classifiers can be solved using so called class balanced measures. The class balanced measures are generally applicable means, appropriate for reporting performance of binary classifiers on balanced as well as on imbalanced datasets. On the basis of the presented pieces of information, a suggestion for a generally applicable, fully-valued, reporting of binary classifiers performance is given. |
eng |
dc.format |
p. 181-192 |
eng |
dc.language.iso |
eng |
eng |
dc.relation.ispartof |
Scientific Papers of the University of Pardubice - Series D, Faculty of Economics and Administration, volume 24, issue: 3 |
eng |
dc.rights |
Práce není přístupná |
eng |
dc.subject |
machine learning |
eng |
dc.subject |
binary classification |
eng |
dc.subject |
class imbalance problem |
eng |
dc.subject |
performance measures |
eng |
dc.subject |
reporting of results |
eng |
dc.title |
On Reporting performance of binary classifiers |
eng |
dc.title.alternative |
Jak správně publikovat výsledky evaluace binárních klasifikátorů |
cze |
dc.type |
article |
eng |
dc.description.abstract-translated |
Článek se zabývá otázkou jak správně provést evaluaci binárních klasifikátorů. Tato otázka je řešena v kontextu vyváženosti zastoupení tříd v evaluačních datových sadách. Některé z metrik používaných při evaluaci jsou totiž citlivé na složení evaluační datových sad. V případě jejich nevyváženosti poskytují tyto metriky zavádějící výsledky. K zamezení tohoto problému je možné použít vážené metriky, které nejsou ovlivňovány skladbou evaluačních datových sad. Tyto metriky jsou zavedeny a diskutovány v rámci tohoto článku. |
cze |
dc.peerreviewed |
yes |
eng |
dc.publicationstatus |
postprint |
eng |
dc.identifier.obd |
39879740 |
eng |
This item appears in the following Collection(s)
Show simple item record
|
Search DSpace
Browse
-
All of DSpace
-
This Collection
My Account
|