Skip to Main content Skip to Navigation
Conference papers

Concevoir un assistant conversationnel de manière itérative et semi-supervisée avec le clustering interactif

Abstract : The design of a dataset needed to train a chatbot is most often the result of manual and tedious step. To guarantee the efficiency of the annotation, we propose the interactive clustering method, an active learning method based on constraints annotation. It’s an iterative approach, relying on a constrained clustering algorithm and using annotator knowledge to lead clustering. In this paper, we expose the process to design a chatbot with the interactive clustering method.
Complete list of metadata

https://hal.inria.fr/hal-03133060
Contributor : Erwan Schild Connect in order to contact the contributor
Submitted on : Friday, February 5, 2021 - 3:56:17 PM
Last modification on : Saturday, October 16, 2021 - 11:26:06 AM
Long-term archiving on: : Friday, May 7, 2021 - 8:28:20 AM

File

Schild et al. - 2021 - Concevo...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03133060, version 1

Collections

Citation

Erwan Schild, Gautier Durantin, Jean-Charles Lamirel. Concevoir un assistant conversationnel de manière itérative et semi-supervisée avec le clustering interactif. Atelier - Fouille de Textes - Text Mine 2021 - En conjonction avec EGC 2021, Association EGC, Jan 2021, Montpellier / Virtual, France. ⟨hal-03133060⟩

Share

Metrics

Les métriques sont temporairement indisponibles