A Methodology to Design Human-Like Embodied Conversational Agents

Abstract : This paper presents a bottom-up methodology to study human-human and human-agent dialogues in order to improve the design of embodied conversational agents (ECAs). The methodology uses a matrix representation of dialogues, constructed by means of a (semi-)automatic annotation process. A multidimensional dialogue pattern extraction and a clustering algorithm are applied to the coding matrices. The extracted patterns model the human behaviours.
Liste complète des métadonnées

Cited literature [55 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00927488
Contributor : Guillaume Dubuisson Duplessis <>
Submitted on : Monday, January 13, 2014 - 10:44:19 AM
Last modification on : Thursday, February 7, 2019 - 5:37:33 PM
Document(s) archivé(s) le : Saturday, April 8, 2017 - 2:38:52 PM

File

2012-dubuisson_duplessis-A-Met...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00927488, version 1

Citation

Zacharie Alès, Guillaume Dubuisson Duplessis, Ovidiu Şerban, Alexandre Pauchet. A Methodology to Design Human-Like Embodied Conversational Agents. International Workshop on Human-Agent Interaction Design and Models (HAIDM'12), 2012, Valencia, Spain. online proceedings. ⟨hal-00927488⟩

Share

Metrics

Record views

226

Files downloads

329