An Extensible and Reusable Pipeline for Automated Utterance Paraphrases - Archive ouverte HAL Access content directly
Conference Papers Lecture Notes in Computer Science Year : 2021

An Extensible and Reusable Pipeline for Automated Utterance Paraphrases

Abstract

In this demonstration paper we showcase an extensible and reusable pipeline for automatic paraphrase generation, i.e., reformulating sentences using different words. Capturing the nuances of human language is fundamental to the effectiveness of Conversational AI systems, as it allows them to deal with the different ways users can utter their requests in natural language. Traditional approaches to utterance paraphrasing acquisition, such as hiring experts or crowd-sourcing, involve processes that are often costly or time consuming, and with their own trade-offs in terms of quality. Automatic paraphrasing is emerging as an attractive alternative that promises a fast, scalable and cost-effective process. In this paper we showcase how our extensible and reusable pipeline for automated utterance paraphrasing can support the development of Conversational AI systems by integrating and extending existing techniques under an unified and configurable framework.
Fichier principal
Vignette du fichier
vldb_paper.pdf (839.47 Ko) Télécharger le fichier
Origin : Publisher files allowed on an open archive

Dates and versions

hal-03399085 , version 1 (23-10-2021)

Identifiers

Cite

Auday Berro, Mohammad-Ali Yaghub Zade Fard, Marcos Baez, Boualem Benatallah, Khalid Benabdeslem. An Extensible and Reusable Pipeline for Automated Utterance Paraphrases. 47th International Conference on Very Large Data Bases, Aug 2021, Copenhagen, Denmark. ⟨10.14778/3476311.3476358⟩. ⟨hal-03399085⟩
120 View
114 Download

Altmetric

Share

Gmail Facebook X LinkedIn More