|

TRINDIKIT is a toolkit for building and
experimenting with dialogue move engines and information
states, that has been developed in the TRINDI and SIRIDUS and TALK projects. TRINDIKIT development is managed by the Dialogue Technology Lab at CLT, the Centre for Language Technology at the University of Gothenburg.
We use
the term information state to mean, roughly, the information stored
internally by an agent, in this case a dialogue system. A
dialogue move engine, or DME, updates the information
state on the basis of observed dialogue moves and selects appropriate
moves to be performed.
Apart from proposing a general system architecture, the TRINDIKIT
also specifies formats for defining information states, update rules,
dialogue moves, and associated algorithms. It further provides a set
of tools for experimenting with different formalizations of
implementations of information states, rules, and algorithms.
To build a dialogue move engine, one needs to provide
definitions of update rules, moves and algorithms, as well as the
internal structure of the information state. One may also add
inference engines, planners, plan recognizers, dialogue grammars,
dialogue game automata etc.. More information on information states
and related concepts is available in Larsson and Traum (2001) and Traum et al. (1999).
|
 |

The DME forms the core of a complete dialogue system. Simple
interpretation, generation, input and output modules are also provided
by the TRINDIKIT, to simulate a end-to-end dialogue system.
Examples of theories and systems implemented using the
TRINDIKIT can be found in Traum et al. (1999) and
Bos et al. (1999).
|
 |