InstanceRank: Bringing order to datasets
In this paper we present InstanceRank, a ranking algorithm that reflects the relevance of the instances within a dataset. InstanceRank applies a similar solution to that used by PageRank, the web pages ranking algorithm in the Google search engine. We also present ISR, an instance selection techniqu...
| Autores: | ; ; |
|---|---|
| Tipo de documento: | Artículo |
| Estado: | Submitted version |
| Fecha de publicación: | 2010 |
| País: | España |
| Institución: | Universidad de Sevilla (US) |
| Repositorio: | idUS. Depósito de Investigación de la Universidad de Sevilla |
| OAI Identifier: | oai:idus.us.es:11441/100070 |
| Acceso en línea: | https://hdl.handle.net/11441/100070 https://doi.org/10.1016/j.patrec.2009.09.022 |
| Nivel de acceso: | Acceso abierto |
| Palabra clave: | Instance-based learning Instance reduction Nearest neighbor PageRank Classification |
| _version_ | 1878730780435283968 |
|---|---|
| author | García Vallejo, Carlos Antonio |
| author2 | Troyano Jiménez, José Antonio Ortega Rodríguez, Francisco Javier |
| author2_role | author author |
| author_browse | García Vallejo, Carlos Antonio Ortega Rodríguez, Francisco Javier Troyano Jiménez, José Antonio |
| author_facet | García Vallejo, Carlos Antonio Troyano Jiménez, José Antonio Ortega Rodríguez, Francisco Javier |
| author_role | author |
| collection | idUS. Depósito de Investigación de la Universidad de Sevilla |
| description | In this paper we present InstanceRank, a ranking algorithm that reflects the relevance of the instances within a dataset. InstanceRank applies a similar solution to that used by PageRank, the web pages ranking algorithm in the Google search engine. We also present ISR, an instance selection technique that uses InstanceRank. This algorithm chooses the most representative instances from a learning database. Experiments show that ISR algorithm, with InstanceRank as ranking criteria, obtains similar results in accuracy to other instance reduction techniques, noticeably reducing the size of the instance set. |
| eu_rights_str_mv | openAccess |
| format | article |
| id | ES_233f8ff35495b70d15bf4e00a41e8b5c |
| institution | Universidad de Sevilla (US) |
| instname_str | Universidad de Sevilla (US) |
| network_acronym_str | ES |
| network_name_str | España |
| oai_identifier_str | oai:idus.us.es:11441/100070 |
| publishDate | 2010 |
| publishDateSort | 2010 |
| publisher | Elsevier |
| publisherStr | Elsevier |
| reponame_str | idUS. Depósito de Investigación de la Universidad de Sevilla |
| spelling | InstanceRank: Bringing order to datasets García Vallejo, Carlos Antonio Troyano Jiménez, José Antonio Ortega Rodríguez, Francisco Javier Instance-based learning Instance reduction Nearest neighbor PageRank Classification In this paper we present InstanceRank, a ranking algorithm that reflects the relevance of the instances within a dataset. InstanceRank applies a similar solution to that used by PageRank, the web pages ranking algorithm in the Google search engine. We also present ISR, an instance selection technique that uses InstanceRank. This algorithm chooses the most representative instances from a learning database. Experiments show that ISR algorithm, with InstanceRank as ranking criteria, obtains similar results in accuracy to other instance reduction techniques, noticeably reducing the size of the instance set. Ministerio de Educación y Ciencia HUM2007-66607-C04-04 Elsevier https://hdl.handle.net/11441/100070 https://doi.org/10.1016/j.patrec.2009.09.022 |
| spellingShingle | InstanceRank: Bringing order to datasets García Vallejo, Carlos Antonio Instance-based learning Instance reduction Nearest neighbor PageRank Classification |
| status_str | submittedVersion |
| title | InstanceRank: Bringing order to datasets |
| title_full | InstanceRank: Bringing order to datasets |
| title_fullStr | InstanceRank: Bringing order to datasets |
| title_full_unstemmed | InstanceRank: Bringing order to datasets |
| title_short | InstanceRank: Bringing order to datasets |
| title_sort | InstanceRank: Bringing order to datasets |
| topic | Instance-based learning Instance reduction Nearest neighbor PageRank Classification |
| topic_facet | Instance-based learning Instance reduction Nearest neighbor PageRank Classification |
| url | https://hdl.handle.net/11441/100070 https://doi.org/10.1016/j.patrec.2009.09.022 |