Evaluation of Different Strategies for Domain Adaptation in Opinion Mining
International audience The work presented in this article takes place in the field of opinion mining and aims more particularly at finding the polarity of a text by relying on machine learning methods. In this context, it focuses on studying various strategies for adapting a statistical classifier t...
Main Authors: | , , |
---|---|
Other Authors: | , , , , , , , , , , , , |
Format: | Conference Object |
Language: | English |
Published: |
HAL CCSD
2014
|
Subjects: | |
Online Access: | https://hal.science/hal-01489996 https://hal.science/hal-01489996/document https://hal.science/hal-01489996/file/LREC2014.pdf |
id |
ftunivparis3:oai:HAL:hal-01489996v1 |
---|---|
record_format |
openpolar |
spelling |
ftunivparis3:oai:HAL:hal-01489996v1 2024-04-14T08:13:42+00:00 Evaluation of Different Strategies for Domain Adaptation in Opinion Mining Garcia-Fernandez, Anne Ferret, Olivier Dinarelli, Marco Laboratoire d'anthropologie sociale (LAS) École des hautes études en sciences sociales (EHESS)-Collège de France (CdF (institution))-Centre National de la Recherche Scientifique (CNRS) Département Intelligence Ambiante et Systèmes Interactifs (DIASI (CEA, LIST)) Laboratoire d'Intégration des Systèmes et des Technologies (LIST (CEA)) Direction de Recherche Technologique (CEA) (DRT (CEA)) Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Direction de Recherche Technologique (CEA) (DRT (CEA)) Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Paris-Saclay École normale supérieure - Paris (ENS-PSL) Université Paris sciences et lettres (PSL) Université Sorbonne Paris Cité (USPC) Lattice - Langues, Textes, Traitements informatiques, Cognition - UMR 8094 (Lattice) Université Sorbonne Nouvelle - Paris 3-Université Sorbonne Paris Cité (USPC)-Centre National de la Recherche Scientifique (CNRS)-Université Paris sciences et lettres (PSL)-Département Littératures et langage - ENS Paris (LILA) Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS-PSL) Reykjavik, Iceland 2014-05-26 https://hal.science/hal-01489996 https://hal.science/hal-01489996/document https://hal.science/hal-01489996/file/LREC2014.pdf en eng HAL CCSD hal-01489996 https://hal.science/hal-01489996 https://hal.science/hal-01489996/document https://hal.science/hal-01489996/file/LREC2014.pdf info:eu-repo/semantics/OpenAccess LREC 2014 proceedings Language Resources Evaluation Conference (LREC) https://hal.science/hal-01489996 Language Resources Evaluation Conference (LREC), May 2014, Reykjavik, Iceland Opinion mining domain adaptation self-training [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL] [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] info:eu-repo/semantics/conferenceObject Conference papers 2014 ftunivparis3 2024-03-21T16:07:50Z International audience The work presented in this article takes place in the field of opinion mining and aims more particularly at finding the polarity of a text by relying on machine learning methods. In this context, it focuses on studying various strategies for adapting a statistical classifier to a new domain when training data only exist for one or several other domains. This study shows more precisely that a self-training procedure consisting in enlarging the initial training corpus with unannotated texts from the target domain that were reliably classified by the classifier is the most successful and stable strategy for the tested domains. Moreover, this strategy gets better results in most cases than (Blitzer et al., 2007)'s method on the same evaluation corpus while it is more simple. Conference Object Iceland Université Sorbonne Nouvelle - Paris 3: HAL |
institution |
Open Polar |
collection |
Université Sorbonne Nouvelle - Paris 3: HAL |
op_collection_id |
ftunivparis3 |
language |
English |
topic |
Opinion mining domain adaptation self-training [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL] [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] |
spellingShingle |
Opinion mining domain adaptation self-training [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL] [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] Garcia-Fernandez, Anne Ferret, Olivier Dinarelli, Marco Evaluation of Different Strategies for Domain Adaptation in Opinion Mining |
topic_facet |
Opinion mining domain adaptation self-training [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL] [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] |
description |
International audience The work presented in this article takes place in the field of opinion mining and aims more particularly at finding the polarity of a text by relying on machine learning methods. In this context, it focuses on studying various strategies for adapting a statistical classifier to a new domain when training data only exist for one or several other domains. This study shows more precisely that a self-training procedure consisting in enlarging the initial training corpus with unannotated texts from the target domain that were reliably classified by the classifier is the most successful and stable strategy for the tested domains. Moreover, this strategy gets better results in most cases than (Blitzer et al., 2007)'s method on the same evaluation corpus while it is more simple. |
author2 |
Laboratoire d'anthropologie sociale (LAS) École des hautes études en sciences sociales (EHESS)-Collège de France (CdF (institution))-Centre National de la Recherche Scientifique (CNRS) Département Intelligence Ambiante et Systèmes Interactifs (DIASI (CEA, LIST)) Laboratoire d'Intégration des Systèmes et des Technologies (LIST (CEA)) Direction de Recherche Technologique (CEA) (DRT (CEA)) Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Direction de Recherche Technologique (CEA) (DRT (CEA)) Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Paris-Saclay École normale supérieure - Paris (ENS-PSL) Université Paris sciences et lettres (PSL) Université Sorbonne Paris Cité (USPC) Lattice - Langues, Textes, Traitements informatiques, Cognition - UMR 8094 (Lattice) Université Sorbonne Nouvelle - Paris 3-Université Sorbonne Paris Cité (USPC)-Centre National de la Recherche Scientifique (CNRS)-Université Paris sciences et lettres (PSL)-Département Littératures et langage - ENS Paris (LILA) Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS-PSL) |
format |
Conference Object |
author |
Garcia-Fernandez, Anne Ferret, Olivier Dinarelli, Marco |
author_facet |
Garcia-Fernandez, Anne Ferret, Olivier Dinarelli, Marco |
author_sort |
Garcia-Fernandez, Anne |
title |
Evaluation of Different Strategies for Domain Adaptation in Opinion Mining |
title_short |
Evaluation of Different Strategies for Domain Adaptation in Opinion Mining |
title_full |
Evaluation of Different Strategies for Domain Adaptation in Opinion Mining |
title_fullStr |
Evaluation of Different Strategies for Domain Adaptation in Opinion Mining |
title_full_unstemmed |
Evaluation of Different Strategies for Domain Adaptation in Opinion Mining |
title_sort |
evaluation of different strategies for domain adaptation in opinion mining |
publisher |
HAL CCSD |
publishDate |
2014 |
url |
https://hal.science/hal-01489996 https://hal.science/hal-01489996/document https://hal.science/hal-01489996/file/LREC2014.pdf |
op_coverage |
Reykjavik, Iceland |
genre |
Iceland |
genre_facet |
Iceland |
op_source |
LREC 2014 proceedings Language Resources Evaluation Conference (LREC) https://hal.science/hal-01489996 Language Resources Evaluation Conference (LREC), May 2014, Reykjavik, Iceland |
op_relation |
hal-01489996 https://hal.science/hal-01489996 https://hal.science/hal-01489996/document https://hal.science/hal-01489996/file/LREC2014.pdf |
op_rights |
info:eu-repo/semantics/OpenAccess |
_version_ |
1796311749035556864 |