Effect of label noise in the complexity of classification problems

Effect of label noise in the complexity of classification problems

Author Garcia, Luis P. F. Google Scholar
Carvalho, Andre C. P. L. F. de Google Scholar
Lorena, Ana C. Autor UNIFESP Google Scholar
Institution Universidade de São Paulo (USP)
Universidade Federal de São Paulo (UNIFESP)
Abstract Noisy data are common in real-World problems and may have several causes, like inaccuracies, distortions or contamination during data collection, storage and/or transmission. the presence of noise in data can affect the complexity of classification problems, making the discrimination of objects from different classes more difficult, and requiring more complex decision boundaries for data separation. in this paper, we investigate how noise affects the complexity of classification problems, by monitoring the sensitivity of several indices of data complexity in the presence of different label noise levels. To characterize the complexity of a classification dataset, we use geometric, statistical and structural measures extracted from data. the experimental results show that some measures are more sensitive than others to the addition of noise in a dataset These measures can be used in the development of new preprocessing techniques for noise identification and novel label noise tolerant algorithms. We thereby show preliminary results on a new filter for noise identification, which is based on two of the complexity measures which were more sensitive to the presence of label noise. (C) 2015 Elsevier B.V. All rights reserved.
Keywords Classification
Label noise
Complexity measures
Noise Filter
Language English
Sponsor Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
Date 2015-07-21
Published in Neurocomputing. Amsterdam: Elsevier B.V., v. 160, p. 108-119, 2015.
ISSN 0925-2312 (Sherpa/Romeo, impact factor)
Publisher Elsevier B.V.
Extent 108-119
Origin http://dx.doi.org/10.1016/j.neucom.2014.10.085
Access rights Closed access
Type Article
Web of Science ID WOS:000354139100010
URI http://repositorio.unifesp.br/handle/11600/39154

Show full item record




File

File Size Format View

There are no files associated with this item.

This item appears in the following Collection(s)

Search


Browse

Statistics

My Account