A multilingual neural coaching model with enhanced long-term dialogue structure

In this work we develop a fully data-driven conversational agent capable of carrying out motivational coach- ing sessions in Spanish, French, Norwegian, and English. Unlike the majority of coaching, and in general well-being related conversational agents that can be found in the literature, ours is...

Descripción completa

Detalles Bibliográficos
Autores: López Zorrilla, Asier, Torres Barañano, María Inés
Tipo de recurso: artículo
Fecha de publicación:2022
País:España
Institución:Universidad del País Vasco
Repositorio:Addi. Archivo Digital para la Docencia y la Investigación
OAI Identifier:oai:addi.ehu.eus:10810/59423
Acceso en línea:http://hdl.handle.net/10810/59423
Access Level:acceso abierto
Palabra clave:dialogue system
coaching
multilingual
transfer learning
explainable artificial intelligence
id ES_bf7ea1a877a76d2c3e500b371f749fe4
oai_identifier_str oai:addi.ehu.eus:10810/59423
network_acronym_str ES
network_name_str España
repository_id_str
spelling A multilingual neural coaching model with enhanced long-term dialogue structureLópez Zorrilla, AsierTorres Barañano, María Inésdialogue systemcoachingmultilingualtransfer learningexplainable artificial intelligenceIn this work we develop a fully data-driven conversational agent capable of carrying out motivational coach- ing sessions in Spanish, French, Norwegian, and English. Unlike the majority of coaching, and in general well-being related conversational agents that can be found in the literature, ours is not designed by hand- crafted rules. Instead, we directly model the coaching strategy of professionals with end users. To this end, we gather a set of virtual coaching sessions through a Wizard of Oz platform, and apply state of the art Natural Language Processing techniques. We employ a transfer learning approach, pretraining GPT2 neural language models and fine-tuning them on our corpus. However, since these only take as input a local dialogue history, a simple fine-tuning procedure is not capable of modeling the long-term dialogue strategies that appear in coaching sessions. To alleviate this issue, we first propose to learn dialogue phase and scenario embeddings in the fine-tuning stage. These indicate to the model at which part of the dialogue it is and which kind of coaching session it is carrying out. Second, we develop a global deep learning system which controls the long-term structure of the dialogue. We also show that this global module can be used to visualize and interpret the decisions taken by the the conversational agent, and that the learnt representations are comparable to dialogue acts. Automatic and human evaluation show that our proposals serve to improve the baseline models. Finally, interaction experiments with coaching experts indicate that the system is usable and gives rise to positive emotions in Spanish, French and English, while the results in Norwegian point out that there is still work to be done in fully data driven approaches with very low resource languages.This work has been partially funded by the Basque Government under grant PRE_2017_1_0357 and by the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 769872.ACMEuropean Commission202320232022info:eu-repo/semantics/articleapplication/pdfhttp://hdl.handle.net/10810/59423reponame:Addi. Archivo Digital para la Docencia y la Investigacióninstname:Universidad del País VascoInglésinfo:eu-repo/grantAgreement/EC/H2020/769872https://dl.acm.org/doi/full/10.1145/3487066info:eu-repo/semantics/openAccess(c) 2022 ACMoai:addi.ehu.eus:10810/594232026-06-18T09:23:17Z
dc.title.none.fl_str_mv A multilingual neural coaching model with enhanced long-term dialogue structure
title A multilingual neural coaching model with enhanced long-term dialogue structure
spellingShingle A multilingual neural coaching model with enhanced long-term dialogue structure
López Zorrilla, Asier
dialogue system
coaching
multilingual
transfer learning
explainable artificial intelligence
title_short A multilingual neural coaching model with enhanced long-term dialogue structure
title_full A multilingual neural coaching model with enhanced long-term dialogue structure
title_fullStr A multilingual neural coaching model with enhanced long-term dialogue structure
title_full_unstemmed A multilingual neural coaching model with enhanced long-term dialogue structure
title_sort A multilingual neural coaching model with enhanced long-term dialogue structure
dc.creator.none.fl_str_mv López Zorrilla, Asier
Torres Barañano, María Inés
author López Zorrilla, Asier
author_facet López Zorrilla, Asier
Torres Barañano, María Inés
author_role author
author2 Torres Barañano, María Inés
author2_role author
dc.contributor.none.fl_str_mv European Commission
dc.subject.none.fl_str_mv dialogue system
coaching
multilingual
transfer learning
explainable artificial intelligence
topic dialogue system
coaching
multilingual
transfer learning
explainable artificial intelligence
description In this work we develop a fully data-driven conversational agent capable of carrying out motivational coach- ing sessions in Spanish, French, Norwegian, and English. Unlike the majority of coaching, and in general well-being related conversational agents that can be found in the literature, ours is not designed by hand- crafted rules. Instead, we directly model the coaching strategy of professionals with end users. To this end, we gather a set of virtual coaching sessions through a Wizard of Oz platform, and apply state of the art Natural Language Processing techniques. We employ a transfer learning approach, pretraining GPT2 neural language models and fine-tuning them on our corpus. However, since these only take as input a local dialogue history, a simple fine-tuning procedure is not capable of modeling the long-term dialogue strategies that appear in coaching sessions. To alleviate this issue, we first propose to learn dialogue phase and scenario embeddings in the fine-tuning stage. These indicate to the model at which part of the dialogue it is and which kind of coaching session it is carrying out. Second, we develop a global deep learning system which controls the long-term structure of the dialogue. We also show that this global module can be used to visualize and interpret the decisions taken by the the conversational agent, and that the learnt representations are comparable to dialogue acts. Automatic and human evaluation show that our proposals serve to improve the baseline models. Finally, interaction experiments with coaching experts indicate that the system is usable and gives rise to positive emotions in Spanish, French and English, while the results in Norwegian point out that there is still work to be done in fully data driven approaches with very low resource languages.
publishDate 2022
dc.date.none.fl_str_mv 2022
2023
2023
dc.type.none.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv http://hdl.handle.net/10810/59423
url http://hdl.handle.net/10810/59423
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv info:eu-repo/grantAgreement/EC/H2020/769872
https://dl.acm.org/doi/full/10.1145/3487066
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
(c) 2022 ACM
eu_rights_str_mv openAccess
rights_invalid_str_mv (c) 2022 ACM
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv ACM
publisher.none.fl_str_mv ACM
dc.source.none.fl_str_mv reponame:Addi. Archivo Digital para la Docencia y la Investigación
instname:Universidad del País Vasco
instname_str Universidad del País Vasco
reponame_str Addi. Archivo Digital para la Docencia y la Investigación
collection Addi. Archivo Digital para la Docencia y la Investigación
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869418396113698816
score 15,301603

Similares en LA Referencia