A plan-based agent architecture for interpreting natural language dialogue
✍ Scribed by LILIANA ARDISSONO; GUIDO BOELLA; LEONARDO LESMO
- Publisher
- Elsevier Science
- Year
- 2000
- Tongue
- English
- Weight
- 523 KB
- Volume
- 52
- Category
- Article
- ISSN
- 1071-5819
No coin nor oath required. For personal study only.
✦ Synopsis
This paper describes a plan-based agent architecture for modelling NL cooperative dialogue; in particular, the paper focuses on the interpretation of dialogue and on the explanation of its coherence by means of the recognition of the speakers' underlying intentions. The approach we propose makes it possible to analyze an explain in a uniform way several apparently unrelated linguistic phenomena, which have been often studied separately and treated via ad-hoc methods in the models of dialogue presented in the literature. Our model of linguistic interaction is based on the idea that dialogue can be seen as any other interaction among agents: therefore, domain-level and linguistic actions are treated in a similar way.
Our agent architecture is based on a two-level representation of the knowledge about acting: at the metalevel, the agent modelling (AM) plans describe the recipes for plan formation and execution (they are a declarative representation of a reactive planner); at the object level, the domain and communicative actions are de"ned. The AM plans are used to identify the goals underlying the actions performed by an observed agent; the recognized plans constitute the dialogue context, where the intentions of all participants are stored in a structured way, in order to be used in the interpretation of the subsequent dialogue turns.
2000 Academic Press KEYWORDS: dialogue processing; plan and goal recognition; agent modelling; natural language interpretation.
A PLAN-BASED AGENT ARCHITECTURE
* A role name.? * An internal variable associated with the role. * The type restriction on the role "llers. E The type declaration of other variables occurring in the action (if any). E The following applicability conditions. * =h-restrictions: they constrain the relations holding among the role "llers. For instance, the domain-level restriction &&has-o$ce(teacher, o$ce'') constraints the two variables to take values such that &&o.ce'' denotes the o$ce of &&teacher''.
* Constraints: they are conditions that specify in which contexts the action can be executed; like the wh-restrictions, constraints cannot be made true if they do not hold, i.e. they cannot be adopted as sub-goals by an agent.
* Preconditions: they are conditions of actions like constraints, but if the agent believes they are false, he can adopt them as sub-goals and try to satisfy them. E The e!ects of the action, i.e. the conditions that may become true after it is executed successfully. If an action execution fails, its e!ects do not a!ect the world state.