A New Method for Vertical Parallelisation of TAN Learning Based on Balanced Incomplete Block Designs
The framework of Bayesian networks is a widely popular formalism for performing belief update under uncertainty. Structure re- stricted Bayesian network models such as the Naive Bayes Model and Tree-Augmented Naive Bayes (TAN) Model have shown impressive per- formance for solving classi cation t...
| Main Authors: | , , , , , |
|---|---|
| Format: | info:eu-repo/semantics/article |
| Language: | English |
| Published: |
2017
|
| Online Access: | http://hdl.handle.net/10835/4857 |
| _version_ | 1789406454215606272 |
|---|---|
| author | Madsen, Anders L. Jensen, Frank Salmerón Cerdán, Antonio Karlsen, Martin Langseth, Helge Nielsen, Thomas D. |
| author_facet | Madsen, Anders L. Jensen, Frank Salmerón Cerdán, Antonio Karlsen, Martin Langseth, Helge Nielsen, Thomas D. |
| author_sort | Madsen, Anders L. |
| collection | DSpace |
| description | The framework of Bayesian networks is a widely popular
formalism for performing belief update under uncertainty. Structure re-
stricted Bayesian network models such as the Naive Bayes Model and
Tree-Augmented Naive Bayes (TAN) Model have shown impressive per-
formance for solving classi cation tasks. However, if the number of vari-
ables or the amount of data is large, then learning a TAN model from
data can be a time consuming task. In this paper, we introduce a new
method for parallel learning of a TAN model from large data sets. The
method is based on computing the mutual information scores between
pairs of variables given the class variable in parallel. The computations
are organised in parallel using balanced incomplete block designs. The
results of a preliminary empirical evaluation of the proposed method on
large data sets show that a signi cant performance improvement is pos-
sible through parallelisation using the method presented in this paper. |
| format | info:eu-repo/semantics/article |
| id | oai:repositorio.ual.es:10835-4857 |
| institution | Universidad de Cuenca |
| language | English |
| publishDate | 2017 |
| record_format | dspace |
| spelling | oai:repositorio.ual.es:10835-48572023-04-12T19:39:59Z A New Method for Vertical Parallelisation of TAN Learning Based on Balanced Incomplete Block Designs Madsen, Anders L. Jensen, Frank Salmerón Cerdán, Antonio Karlsen, Martin Langseth, Helge Nielsen, Thomas D. The framework of Bayesian networks is a widely popular formalism for performing belief update under uncertainty. Structure re- stricted Bayesian network models such as the Naive Bayes Model and Tree-Augmented Naive Bayes (TAN) Model have shown impressive per- formance for solving classi cation tasks. However, if the number of vari- ables or the amount of data is large, then learning a TAN model from data can be a time consuming task. In this paper, we introduce a new method for parallel learning of a TAN model from large data sets. The method is based on computing the mutual information scores between pairs of variables given the class variable in parallel. The computations are organised in parallel using balanced incomplete block designs. The results of a preliminary empirical evaluation of the proposed method on large data sets show that a signi cant performance improvement is pos- sible through parallelisation using the method presented in this paper. 2017-06-14T09:55:57Z 2017-06-14T09:55:57Z 2014 info:eu-repo/semantics/article http://hdl.handle.net/10835/4857 en Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ info:eu-repo/semantics/openAccess |
| spellingShingle | Madsen, Anders L. Jensen, Frank Salmerón Cerdán, Antonio Karlsen, Martin Langseth, Helge Nielsen, Thomas D. A New Method for Vertical Parallelisation of TAN Learning Based on Balanced Incomplete Block Designs |
| title | A New Method for Vertical Parallelisation of TAN Learning Based on Balanced Incomplete Block Designs |
| title_full | A New Method for Vertical Parallelisation of TAN Learning Based on Balanced Incomplete Block Designs |
| title_fullStr | A New Method for Vertical Parallelisation of TAN Learning Based on Balanced Incomplete Block Designs |
| title_full_unstemmed | A New Method for Vertical Parallelisation of TAN Learning Based on Balanced Incomplete Block Designs |
| title_short | A New Method for Vertical Parallelisation of TAN Learning Based on Balanced Incomplete Block Designs |
| title_sort | new method for vertical parallelisation of tan learning based on balanced incomplete block designs |
| url | http://hdl.handle.net/10835/4857 |
| work_keys_str_mv | AT madsenandersl anewmethodforverticalparallelisationoftanlearningbasedonbalancedincompleteblockdesigns AT jensenfrank anewmethodforverticalparallelisationoftanlearningbasedonbalancedincompleteblockdesigns AT salmeroncerdanantonio anewmethodforverticalparallelisationoftanlearningbasedonbalancedincompleteblockdesigns AT karlsenmartin anewmethodforverticalparallelisationoftanlearningbasedonbalancedincompleteblockdesigns AT langsethhelge anewmethodforverticalparallelisationoftanlearningbasedonbalancedincompleteblockdesigns AT nielsenthomasd anewmethodforverticalparallelisationoftanlearningbasedonbalancedincompleteblockdesigns AT madsenandersl newmethodforverticalparallelisationoftanlearningbasedonbalancedincompleteblockdesigns AT jensenfrank newmethodforverticalparallelisationoftanlearningbasedonbalancedincompleteblockdesigns AT salmeroncerdanantonio newmethodforverticalparallelisationoftanlearningbasedonbalancedincompleteblockdesigns AT karlsenmartin newmethodforverticalparallelisationoftanlearningbasedonbalancedincompleteblockdesigns AT langsethhelge newmethodforverticalparallelisationoftanlearningbasedonbalancedincompleteblockdesigns AT nielsenthomasd newmethodforverticalparallelisationoftanlearningbasedonbalancedincompleteblockdesigns |