5th International Conference on Spoken Language Processing

Sydney, Australia
November 30 - December 4, 1998

Automatic Classification of Dialogue Contexts for Dialogue Predictions

Cosmin Popovici (1), Paolo Baggia (2), Pietro Laface (3), Loreta Moisa (3)

(1) I.C.I. - Institutul de Cercetari in Informatica, Romania
(2) CSELT - Centro Studi e Laboratori Telecomunicazioni, Italy
(3) Politecnico di Torino, Italy

This paper exploits the broad concept of dialogue predictions by linking a point in a human-machine dialogue with a specific language model which is used during the recognition of the next user utterance. The idea is to cluster several dialogue contexts into a class and to create for each class a specific language model. We present an automatic algorithm based on the minimal decrease of mutual information which clusters the dialogue contexts. Moreover the algorithm is able to guess an appropriate number of classes, that gives a good trade off between the mutual information and the amount of training data. Therefore the automatic classification allows the full automatic creation of context-dependent language models for a spoken dialogue system.

Full Paper

Bibliographic reference.  Popovici, Cosmin / Baggia, Paolo / Laface, Pietro / Moisa, Loreta (1998): "Automatic classification of dialogue contexts for dialogue predictions", In ICSLP-1998, paper 0552.