Active Learning and Multi-label Classification for Ellipsis and Coreference Detection in Conversational Question-Answering - Archive ouverte HAL Access content directly
Conference Papers Year : 2021

Active Learning and Multi-label Classification for Ellipsis and Coreference Detection in Conversational Question-Answering

Quentin Brabant
Lina Maria Rojas-Barahona

Abstract

In human conversations, ellipsis and coreference are commonly occurring linguistic phenomena. Although these phenomena are a mean of making human-machine conversations more fluent and natural, only few dialogue corpora contain explicit indications on which turns contain ellipses and/or coreferences. In this paper we address the task of automatically detecting ellipsis and coreferences in conversational question answering. We propose to use a multi-label classifier based on DistilBERT. Multi-label classification and active learning are employed to compensate the limited amount of labeled data. We show that these methods greatly enhance the performance of the classifier for detecting these phenomena on a manually labeled dataset.
Fichier principal
Vignette du fichier
EMNLP_workshop_paper.pdf (162.49 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03533906 , version 1 (19-01-2022)

Identifiers

  • HAL Id : hal-03533906 , version 1

Cite

Quentin Brabant, Lina Maria Rojas-Barahona, Claire Gardent. Active Learning and Multi-label Classification for Ellipsis and Coreference Detection in Conversational Question-Answering. 12th International Workshop on Spoken Dialog System Technology (IWSDS 2021), Nov 2021, Singapour/Virtual, Singapore. ⟨hal-03533906⟩
43 View
102 Download

Share

Gmail Mastodon Facebook X LinkedIn More