On positive and unlabeled learning for text classification

István Nagy T., Richárd Farkas, János Csirik

Research output: Conference contribution

4 Citations (Scopus)

Abstract

In this paper we present a slightly modified machine learning approach for text classification working exclusively from positive and unlabeled samples. Our method can assure that the positive class is not underrepresented during the iterative training process and it can achieve 30% better F-value when the amount of positive examples is low.

Original languageEnglish
Title of host publicationText, Speech and Dialogue - 14th International Conference, TSD 2011, Proceedings
Pages219-226
Number of pages8
DOIs
Publication statusPublished - szept. 19 2011
Event14th International Conference on Text, Speech and Dialogue, TSD 2011 - Pilsen, Czech Republic
Duration: szept. 1 2011szept. 5 2011

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6836 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other14th International Conference on Text, Speech and Dialogue, TSD 2011
CountryCzech Republic
CityPilsen
Period9/1/119/5/11

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'On positive and unlabeled learning for text classification'. Together they form a unique fingerprint.

  • Cite this

    Nagy T., I., Farkas, R., & Csirik, J. (2011). On positive and unlabeled learning for text classification. In Text, Speech and Dialogue - 14th International Conference, TSD 2011, Proceedings (pp. 219-226). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6836 LNAI). https://doi.org/10.1007/978-3-642-23538-2_28