Utilize este identificador para referenciar este registo:
https://hdl.handle.net/10216/43258Registo completo
| Campo DC | Valor | Idioma |
|---|---|---|
| dc.creator | Luís Sarmento | |
| dc.creator | Sérgio Nunes | |
| dc.creator | Jorge Teixeira | |
| dc.creator | Eugénio Oliveira | |
| dc.date.accessioned | 2022-09-10T18:11:19Z | - |
| dc.date.available | 2022-09-10T18:11:19Z | - |
| dc.date.issued | 2009 | |
| dc.identifier.other | sigarra:57576 | |
| dc.identifier.uri | https://hdl.handle.net/10216/43258 | - |
| dc.description.abstract | We propose an unsupervised method for propagating automatically extracted fine-grained topic labels among news items to improve their topic description for subsequent text classification procedure. This method compares vector representations of news items and assigns to each news item the label of its closest neighbour with a different topic label. Results obtained show that high precision can be achieved in propagating the top ranked topic label, and that 2-gram and 3-gram feature representations optimize the precision. | |
| dc.language.iso | eng | |
| dc.relation.ispartof | 2009 IEEE/WIC/ACM INTERNATIONAL JOINT CONFERENCES ON WEB INTELLIGENCE (WI) AND INTELLIGENT AGENT TECHNOLOGIES (IAT), VOL 2 | |
| dc.rights | openAccess | |
| dc.rights.uri | https://creativecommons.org/licenses/by-nc/4.0/ | |
| dc.subject | Inteligência artificial, Ciências da computação e da informação | |
| dc.subject | Artificial intelligence, Computer and information sciences | |
| dc.title | Propagating fine-grained topic labels in news snippets | |
| dc.type | Artigo em Livro de Atas de Conferência Internacional | |
| dc.contributor.uporto | Faculdade de Engenharia | |
| dc.identifier.doi | 10.1109/wi-iat.2009.338 | |
| dc.identifier.authenticus | P-003-SWF | |
| dc.subject.fos | Ciências exactas e naturais::Ciências da computação e da informação | |
| dc.subject.fos | Natural sciences::Computer and information sciences | |
| Aparece nas coleções: | FEUP - Artigo em Livro de Atas de Conferência Internacional | |
Este registo está protegido por Licença Creative Commons
