Modeling concept drift: A probabilistic graphical model based approach
An often used approach for detecting and adapting to concept drift when doing classi cation is to treat the data as i.i.d. and use changes in classi cation accuracy as an indication of concept drift. In this paper, we take a different perspective and propose a framework, based on probabilistic graph...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | info:eu-repo/semantics/article |
| Language: | English |
| Published: |
2017
|
| Online Access: | http://hdl.handle.net/10835/4861 |
| _version_ | 1789406621644881920 |
|---|---|
| author | Borchani, Hanen Martínez, Ana M. Masegosa, Andrés R. Langseth, Helge Nielsen, Thomas D. Salmerón Cerdán, Antonio Fernández, Antonio Madsen, Anders L. Sáez, Ramón |
| author_facet | Borchani, Hanen Martínez, Ana M. Masegosa, Andrés R. Langseth, Helge Nielsen, Thomas D. Salmerón Cerdán, Antonio Fernández, Antonio Madsen, Anders L. Sáez, Ramón |
| author_sort | Borchani, Hanen |
| collection | DSpace |
| description | An often used approach for detecting and adapting to concept drift when doing classi cation is to treat the data as i.i.d. and use changes in classi cation accuracy as an indication of concept drift. In this paper, we take a different perspective and propose a framework, based on probabilistic graphical models, that explicitly represents concept drift using latent variables. To ensure effcient inference and learning, we resort to a variational Bayes inference scheme. As a proof of concept, we demonstrate and analyze the proposed framework using synthetic data sets as well as a real fi nancial data set from a Spanish bank. |
| format | info:eu-repo/semantics/article |
| id | oai:repositorio.ual.es:10835-4861 |
| institution | Universidad de Cuenca |
| language | English |
| publishDate | 2017 |
| record_format | dspace |
| spelling | oai:repositorio.ual.es:10835-48612023-04-12T19:40:20Z Modeling concept drift: A probabilistic graphical model based approach Borchani, Hanen Martínez, Ana M. Masegosa, Andrés R. Langseth, Helge Nielsen, Thomas D. Salmerón Cerdán, Antonio Fernández, Antonio Madsen, Anders L. Sáez, Ramón An often used approach for detecting and adapting to concept drift when doing classi cation is to treat the data as i.i.d. and use changes in classi cation accuracy as an indication of concept drift. In this paper, we take a different perspective and propose a framework, based on probabilistic graphical models, that explicitly represents concept drift using latent variables. To ensure effcient inference and learning, we resort to a variational Bayes inference scheme. As a proof of concept, we demonstrate and analyze the proposed framework using synthetic data sets as well as a real fi nancial data set from a Spanish bank. 2017-06-16T08:22:52Z 2017-06-16T08:22:52Z 2015 info:eu-repo/semantics/article http://hdl.handle.net/10835/4861 en Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ info:eu-repo/semantics/openAccess |
| spellingShingle | Borchani, Hanen Martínez, Ana M. Masegosa, Andrés R. Langseth, Helge Nielsen, Thomas D. Salmerón Cerdán, Antonio Fernández, Antonio Madsen, Anders L. Sáez, Ramón Modeling concept drift: A probabilistic graphical model based approach |
| title | Modeling concept drift: A probabilistic graphical model based approach |
| title_full | Modeling concept drift: A probabilistic graphical model based approach |
| title_fullStr | Modeling concept drift: A probabilistic graphical model based approach |
| title_full_unstemmed | Modeling concept drift: A probabilistic graphical model based approach |
| title_short | Modeling concept drift: A probabilistic graphical model based approach |
| title_sort | modeling concept drift: a probabilistic graphical model based approach |
| url | http://hdl.handle.net/10835/4861 |
| work_keys_str_mv | AT borchanihanen modelingconceptdriftaprobabilisticgraphicalmodelbasedapproach AT martinezanam modelingconceptdriftaprobabilisticgraphicalmodelbasedapproach AT masegosaandresr modelingconceptdriftaprobabilisticgraphicalmodelbasedapproach AT langsethhelge modelingconceptdriftaprobabilisticgraphicalmodelbasedapproach AT nielsenthomasd modelingconceptdriftaprobabilisticgraphicalmodelbasedapproach AT salmeroncerdanantonio modelingconceptdriftaprobabilisticgraphicalmodelbasedapproach AT fernandezantonio modelingconceptdriftaprobabilisticgraphicalmodelbasedapproach AT madsenandersl modelingconceptdriftaprobabilisticgraphicalmodelbasedapproach AT saezramon modelingconceptdriftaprobabilisticgraphicalmodelbasedapproach |