MACA: A Modular Architecture for Conversational Agents

Hoai Phuoc Truong, Prasanna Parthasarathi, Joelle Pineau


Abstract
We propose a software architecture designed to ease the implementation of dialogue systems. The Modular Architecture for Conversational Agents (MACA) uses a plug-n-play style that allows quick prototyping, thereby facilitating the development of new techniques and the reproduction of previous work. The architecture separates the domain of the conversation from the agent’s dialogue strategy, and as such can be easily extended to multiple domains. MACA provides tools to host dialogue agents on Amazon Mechanical Turk (mTurk) for data collection and allows processing of other sources of training data. The current version of the framework already incorporates several domains and existing dialogue strategies from the recent literature.
Anthology ID:
W17-5513
Volume:
Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue
Month:
August
Year:
2017
Address:
Saarbrücken, Germany
Editors:
Kristiina Jokinen, Manfred Stede, David DeVault, Annie Louis
Venue:
SIGDIAL
SIG:
SIGDIAL
Publisher:
Association for Computational Linguistics
Note:
Pages:
93–102
Language:
URL:
https://aclanthology.org/W17-5513
DOI:
10.18653/v1/W17-5513
Bibkey:
Cite (ACL):
Hoai Phuoc Truong, Prasanna Parthasarathi, and Joelle Pineau. 2017. MACA: A Modular Architecture for Conversational Agents. In Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue, pages 93–102, Saarbrücken, Germany. Association for Computational Linguistics.
Cite (Informal):
MACA: A Modular Architecture for Conversational Agents (Truong et al., SIGDIAL 2017)
Copy Citation:
PDF:
https://aclanthology.org/W17-5513.pdf