Improving text classification accuracy by training label cleaning (Articolo in rivista)

Type

• Prodotto della ricerca (Classe)
• Articolo in rivista (Classe)

Label

• Improving text classification accuracy by training label cleaning (Articolo in rivista) (literal)

Anno

• 2013-01-01T00:00:00+01:00 (literal)

Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#doi

• 10.1145/2516889 (literal)

Alternative label

• Esuli A., Sebastiani F. (2013)
  Improving text classification accuracy by training label cleaning
  in ACM transactions on information systems
  (literal)

Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#autori

• Esuli A., Sebastiani F. (literal)

Pagina inizio

• 19 (literal)

Pagina fine

• 28 (literal)

Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#url

• http://dl.acm.org/citation.cfm?doid=2536736.2516889 (literal)

Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#numeroVolume

• 31 (literal)

Rivista

• ACM transactions on information systems (Rivista)

Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#numeroFascicolo

• 4 (literal)

Note

• PuMa (literal)
• ISI Web of Science (WOS) (literal)
• Scopu (literal)

Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#affiliazioni

• CNR-ISTI, Pisa, Italy; CNR-ISTI, Pisa, Italy (literal)

Titolo

• Improving text classification accuracy by training label cleaning (literal)

Abstract

• In text classification (TC) and other tasks involving supervised learning, labelled data may be scarce or expensive to obtain. Semisupervised learning and active learning are two strategies whose aim is maximizing the effectiveness of the resulting classifiers for a given amount of training effort. Both strategies have been actively investigated for TC in recent years. Much less research has been devoted to a third such strategy, training label cleaning (TLC), which consists in devising ranking functions that sort the original training examples in terms of how likely it is that the human annotator has mislabelled them. This provides a convenient means for the human annotator to revise the training set so as to improve its quality. Working in the context of boosting-based learning methods for multilabel classification we present three different techniques for performing TLC and, on three widely used TC benchmarks, evaluate them by their capability of spotting training documents that, for experimental reasons only, we have purposefully mislabelled. We also evaluate the degradation in classification effectiveness that these mislabelled texts bring about, and to what extent training label cleaning can prevent this degradation. (literal)

Prodotto di

• Istituto di scienza e tecnologie dell'informazione \"Alessandro Faedo\" (ISTI) (Istituto)
• Digital Libraries (ICT.P08.010.002) (Modulo)

Autore CNR

• FABRIZIO SEBASTIANI (Unità di personale interno)
• ANDREA ESULI (Unità di personale interno)

Insieme di parole chiave

• Keywords of "Improving text classification accuracy by training label cleaning" (Insieme di parole chiave)

Incoming links:

Prodotto

• Istituto di scienza e tecnologie dell'informazione \"Alessandro Faedo\" (ISTI) (Istituto)
• Digital Libraries (ICT.P08.010.002) (Modulo)

Autore CNR di

• FABRIZIO SEBASTIANI (Unità di personale interno)
• ANDREA ESULI (Unità di personale interno)

Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#rivistaDi

• ACM transactions on information systems (Rivista)

Insieme di parole chiave di

• Keywords of "Improving text classification accuracy by training label cleaning" (Insieme di parole chiave)