Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Comparing user simulations for dialogue strategy learning

Comparing user simulations for dialogue strategy learning Comparing User Simulations for Dialogue Strategy Learning HUA AI and DIANE LITMAN, University of Pittsburgh Recent studies show that user simulations can be used to generate training corpora for learning dialogue strategies automatically. However, it is unclear what type of simulation is most suitable in a particular task setting. We observe that a simulation which generates random behaviors in a restricted way outperforms simulations that mimic human user behaviors in a statistical way. Our nding suggests that we do not always need to construct a realistic user simulation. Since constructing realistic user simulations is not a trivial task, we can save engineering cost by wisely choosing simulation models that are appropriate for our task. Categories and Subject Descriptors: I.2.7 [Arti cial Intelligence]: Natural Language Processing General Terms: Design Additional Key Words and Phrases: Dialogue System, dialogue strategy learning, user simulation ACM Reference Format: Ai, H. and Litman, D. 2011. Comparing user simulations for dialogue strategy learning. ACM Trans. Speech Lang. Process. 7, 3, Article 9 (May 2011), 18 pages. DOI = 10.1145/1966407.1966414 http://doi.acm.org/10.1145/1966407.1966414 1. INTRODUCTION Recent advances in spoken language understanding have made it possible to develop spoken dialogue systems for many applications in information services, entertainment, education, http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png ACM Transactions on Speech and Language Processing (TSLP) Association for Computing Machinery

Comparing user simulations for dialogue strategy learning

Loading next page...
 
/lp/association-for-computing-machinery/comparing-user-simulations-for-dialogue-strategy-learning-dXmCWaDxY2
Publisher
Association for Computing Machinery
Copyright
Copyright © 2011 by ACM Inc.
ISSN
1550-4875
DOI
10.1145/1966407.1966414
Publisher site
See Article on Publisher Site

Abstract

Comparing User Simulations for Dialogue Strategy Learning HUA AI and DIANE LITMAN, University of Pittsburgh Recent studies show that user simulations can be used to generate training corpora for learning dialogue strategies automatically. However, it is unclear what type of simulation is most suitable in a particular task setting. We observe that a simulation which generates random behaviors in a restricted way outperforms simulations that mimic human user behaviors in a statistical way. Our nding suggests that we do not always need to construct a realistic user simulation. Since constructing realistic user simulations is not a trivial task, we can save engineering cost by wisely choosing simulation models that are appropriate for our task. Categories and Subject Descriptors: I.2.7 [Arti cial Intelligence]: Natural Language Processing General Terms: Design Additional Key Words and Phrases: Dialogue System, dialogue strategy learning, user simulation ACM Reference Format: Ai, H. and Litman, D. 2011. Comparing user simulations for dialogue strategy learning. ACM Trans. Speech Lang. Process. 7, 3, Article 9 (May 2011), 18 pages. DOI = 10.1145/1966407.1966414 http://doi.acm.org/10.1145/1966407.1966414 1. INTRODUCTION Recent advances in spoken language understanding have made it possible to develop spoken dialogue systems for many applications in information services, entertainment, education,

Journal

ACM Transactions on Speech and Language Processing (TSLP)Association for Computing Machinery

Published: May 1, 2011

There are no references for this article.