Paper: Modeling Dialogue Structure with Adjacency Pair Analysis and Hidden Markov Models

ACL ID N09-2013
Title Modeling Dialogue Structure with Adjacency Pair Analysis and Hidden Markov Models
Venue Human Language Technologies
Session Short Paper
Year 2009
Authors

Automaticaly detecting dialogue structure within corpora of human-human dialogue is the subject of increasing atention. In the do- main of tutorial dialogue, automatic discovery of dialogue structure is of particular interest because these structures inherently represent tutorial strategies or modes, the study of which is key to the design of inteligent tutor- ing systems that comunicate with learners through natural language. We propose a methodology in which a corpus of human- human tutorial dialogue is first manualy an- notated with dialogue acts. Dependent adja- cency pairs of these acts are then identified through χ 2 analysis, and hiden Markov mod- eling is aplied to the observed sequences to induce a descriptive model of the dialogue structure.