Dialogue knowledge acquisition from annotated corpora

Author

Kita, L. ; Fukui, Yoshikazu ; Nagata, Masaaki ; Morimoto, Tsuyoshi

Author_Institution

Fac. of Eng., Tokushima Univ., Japan

Volume

1

fYear

1996

Firstpage

556

Abstract

In the work described, we automatically deduce dialogue structures from a corpus with probabilistic methods. Each utterance in the corpus is annotated with a speaker label and an utterance type called IFT (illocutionary force type). We use an ergodic HMM (hidden Markov model) and the ALERGIA algorithm, an algorithm for learning probabilistic automata by means of state merging, to model the speaker-IFT sequences. Our experiments successfully extract typical dialogue structures such as turn-taking and speech act sequencing

Keywords

hidden Markov models; knowledge acquisition; natural languages; probabilistic automata; speech processing; ALERGIA algorithm; annotated corpora; dialogue knowledge acquisition; dialogue structures; ergodic hidden Markov model; illocutionary force type; probabilistic automata; probabilistic methods; speaker label; speech act sequencing; state merging; turn-taking; utterance type; Communication systems; Databases; Hidden Markov models; Knowledge acquisition; Laboratories; Learning automata; Merging; Natural language processing; Natural languages; Speech processing;

fLanguage

English

Publisher

ieee

Conference_Titel

Systems, Man, and Cybernetics, 1996., IEEE International Conference on

ISSN

1062-922X

Print_ISBN

0-7803-3280-6

Type

conf

DOI

10.1109/ICSMC.1996.569852

Filename

569852