Publications

Advanced search

Abstract

Vicent Tamarit, Carlos D. Martínez-Hinarejos, José-Miguel Benedí. Improving Unsegmented Statistical Dialogue Act Labelling. Proceedings of the International Conference Recent Advances in Natural Language Processing 2009, 2009. Galia Angelova, Kalina Bontcheva, Ruslan Mitkov, Nikolai Nikolov (Editors). pp. 434-440. RANLP.

An important part of a dialogue system is the correct labelling of turns with dialogue-related meaning. This meaning is usually represented by dialogue acts, which give the system semantic information about user intentions. Each dialogue act gives the semantic of a segment of a turn, which can be formed by several segments. Probabilistic models that perform dialogue act labelling can be used on segmented or unsegmented turns. The last option is the more realistic one, but provides poorer results. An hypothesis on the number of segments can be provided in this case to improve the results. We propose some methods to estimate the probability of the number of segments based on the transcription of the turn. The new labelling model includes the estimation of the probability of the number of segments in the turn. The results show that this inclusion significantly improves the labelling accuracy.