Elastic CRFs for Open-Ontology Slot Filling

Slot filling is a crucial component in task-oriented dialog systems that is used to parse (user) utterances into semantic concepts called slots. An ontology is defined by the collection of slots and the values that each slot can take. The most widely used practice of treating slot filling as a seque...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Yinpei Dai, Yichi Zhang, Hong Liu, Zhijian Ou, Yi Huang, Junlan Feng
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	open ontology slot filling conditional random fields dialog systems Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999
Acceso en línea:	https://doaj.org/article/e5c41e8c86a84008b4b1fc6920030404
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:e5c41e8c86a84008b4b1fc6920030404
record_format	dspace
spelling	oai:doaj.org-article:e5c41e8c86a84008b4b1fc69200304042021-11-25T16:35:00ZElastic CRFs for Open-Ontology Slot Filling10.3390/app1122106752076-3417https://doaj.org/article/e5c41e8c86a84008b4b1fc69200304042021-11-01T00:00:00Zhttps://www.mdpi.com/2076-3417/11/22/10675https://doaj.org/toc/2076-3417Slot filling is a crucial component in task-oriented dialog systems that is used to parse (user) utterances into semantic concepts called slots. An ontology is defined by the collection of slots and the values that each slot can take. The most widely used practice of treating slot filling as a sequence labeling task suffers from two main drawbacks. First, the ontology is usually pre-defined and fixed and therefore is not able to detect new labels for unseen slots. Second, the one-hot encoding of slot labels ignores the correlations between slots with similar semantics, which makes it difficult to share knowledge learned across different domains. To address these problems, we propose a new model called elastic conditional random field (eCRF), where each slot is represented by the embedding of its natural language description and modeled by a CRF layer. New slot values can be detected by eCRF whenever a language description is available for the slot. In our experiment, we show that eCRFs outperform existing models in both in-domain and cross-domain tasks, especially in predicting unseen slots and values.Yinpei DaiYichi ZhangHong LiuZhijian OuYi HuangJunlan FengMDPI AGarticleopen ontologyslot fillingconditional random fieldsdialog systemsTechnologyTEngineering (General). Civil engineering (General)TA1-2040Biology (General)QH301-705.5PhysicsQC1-999ChemistryQD1-999ENApplied Sciences, Vol 11, Iss 10675, p 10675 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	open ontology slot filling conditional random fields dialog systems Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999
spellingShingle	open ontology slot filling conditional random fields dialog systems Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999 Yinpei Dai Yichi Zhang Hong Liu Zhijian Ou Yi Huang Junlan Feng Elastic CRFs for Open-Ontology Slot Filling
description	Slot filling is a crucial component in task-oriented dialog systems that is used to parse (user) utterances into semantic concepts called slots. An ontology is defined by the collection of slots and the values that each slot can take. The most widely used practice of treating slot filling as a sequence labeling task suffers from two main drawbacks. First, the ontology is usually pre-defined and fixed and therefore is not able to detect new labels for unseen slots. Second, the one-hot encoding of slot labels ignores the correlations between slots with similar semantics, which makes it difficult to share knowledge learned across different domains. To address these problems, we propose a new model called elastic conditional random field (eCRF), where each slot is represented by the embedding of its natural language description and modeled by a CRF layer. New slot values can be detected by eCRF whenever a language description is available for the slot. In our experiment, we show that eCRFs outperform existing models in both in-domain and cross-domain tasks, especially in predicting unseen slots and values.
format	article
author	Yinpei Dai Yichi Zhang Hong Liu Zhijian Ou Yi Huang Junlan Feng
author_facet	Yinpei Dai Yichi Zhang Hong Liu Zhijian Ou Yi Huang Junlan Feng
author_sort	Yinpei Dai
title	Elastic CRFs for Open-Ontology Slot Filling
title_short	Elastic CRFs for Open-Ontology Slot Filling
title_full	Elastic CRFs for Open-Ontology Slot Filling
title_fullStr	Elastic CRFs for Open-Ontology Slot Filling
title_full_unstemmed	Elastic CRFs for Open-Ontology Slot Filling
title_sort	elastic crfs for open-ontology slot filling
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/e5c41e8c86a84008b4b1fc6920030404
work_keys_str_mv	AT yinpeidai elasticcrfsforopenontologyslotfilling AT yichizhang elasticcrfsforopenontologyslotfilling AT hongliu elasticcrfsforopenontologyslotfilling AT zhijianou elasticcrfsforopenontologyslotfilling AT yihuang elasticcrfsforopenontologyslotfilling AT junlanfeng elasticcrfsforopenontologyslotfilling
_version_	1718413110138634240

Elastic CRFs for Open-Ontology Slot Filling

Ejemplares similares