ISCA Archive Interspeech 2008
ISCA Archive Interspeech 2008

Effects of user modeling on POMDP-based dialogue systems

Dongho Kim, Hyeong Seop Sim, Kee-Eung Kim, Jin Hyung Kim, Hyunjeong Kim, Joo Won Sung

Partially observable Markov decision processes (POMDPs) have gained significant interest in research on spoken dialogue systems, due to among many benefits its ability to naturally model the dialogue strategy selection problem under the unreliability in automated speech recognition. However, the POMDP approaches are essentially model-based, and as a result, the dialogue strategy computed from POMDP is subject to the correctness of the model. In this paper, we extend some of the previous user models for POMDPs, and evaluate the effects of user models on the dialogue strategy computed from POMDP.


doi: 10.21437/Interspeech.2008-354

Cite as: Kim, D., Sim, H.S., Kim, K.-E., Kim, J.H., Kim, H., Sung, J.W. (2008) Effects of user modeling on POMDP-based dialogue systems. Proc. Interspeech 2008, 1169-1172, doi: 10.21437/Interspeech.2008-354

@inproceedings{kim08b_interspeech,
  author={Dongho Kim and Hyeong Seop Sim and Kee-Eung Kim and Jin Hyung Kim and Hyunjeong Kim and Joo Won Sung},
  title={{Effects of user modeling on POMDP-based dialogue systems}},
  year=2008,
  booktitle={Proc. Interspeech 2008},
  pages={1169--1172},
  doi={10.21437/Interspeech.2008-354}
}