The TRAINS project: a case study in building a conversational planning agent

@article{Allen1995TheTP,
  title={The TRAINS project: a case study in building a conversational planning agent},
  author={James F. Allen and Lenhart K. Schubert and George Ferguson and Peter A. Heeman and Chung Hee Hwang and Tsuneaki Kato and Marc Light and Nathaniel G. Martin and Bradford W. Miller and Massimo Poesio and David R. Traum},
  journal={J. Exp. Theor. Artif. Intell.},
  year={1995},
  volume={7},
  pages={7-48}
}
The TRAINS project is an effort to build a conversationally proficient planning assistant. A key part of the project is the construction of the TRAINS system, which provides the research platform for a wide range of issues in natural language understanding, mixed-initiative planning systems, and representing and reasoning about time, actions and events. Four years have now passed since the beginning of the project. Each year we have produced a demonstration system that focused on a dialog that… 

Figures from this paper

Knowledge Representation in the TRAINS-93 Conversation System

TLDR
It is found that using very expressive representations minimized the tensions, since it is easier to extract what one needs from an elaborate representation retaining all semantic nuances, than to make up for lost information.

Knowledge Representation in the TRAINS-93 Conversation

TLDR
It is found that using very expressive representations minimized the tensions, since it is easier to extract what one needs from an elaborate representation retaining all semantic nuances, than to make up for lost information.

TRAINS as an Embodied Natural Language Dialogue System

TLDR
The TRAINS system is considered as an embodied natural language system that can interact and collaborate with humans in problem solving tasks, combining situation assessment, interactive plan construction, evaluation and execution.

Planning in Complex Worlds via Mixed-Initiative Interaction

TLDR
An overview of research at Rochester addressing problems in developing large-scale plans in complex worlds is presented, viewing the human as an essential part of the planning process and dramatically change the problems that are important for the ultimate successful application of planning technology.

TRAINS-95: Towards a Mixed-Initiative Planning Assistant

TLDR
The implementation of a prototype version of a mixed-initiative planning system, TRAINS-95, which helps a manager solve routing problems in a simple transportation domain, and how traditional planning technology does not play a major role in the system is described.

Knowledge Representation in the TRAINS System

TLDR
The goal of the paper is to describe how TRAINS-93, the latest prototype of TRAINS, deals with representational needs, representing lexical meaning, dealing with the problem of ambiguity, make use of information about context, and finding a connection between the content of the current utterance and the plan being jointly developed by system and user.

Planeamiento en un Sistema de Dialogo

TLDR
This work investigates how to add a planning step to a dialog system FrOz (a computer game) and works with the planner PaDoK developed at the University of Rome that implements state-of-the-art techniques.

Turn planning for a dialogue-based intelligent tutoring system

TLDR
A new method of lexical analysis that allows the researcher to visualize the alternatives conveniently is discussed that is applicable to other natural language dialogue systems, especially Socratic dialogue systems.

Knowledge representation and reasoning for mixed-initiative planning

TLDR
This dissertation describes the formal foundations and implementation of a commonsense, mixed-initiative plan reasoning system, taking as its data human communicative and plan reasoning abilities and developing formalisms that characterize these abilities and systems that approximate them.

TRIPS: An Integrated Intelligent Problem-Solving Assistant

TLDR
How the integrated system provides key advantages for helping both work in natural language dialogue processing and in interactive planning and problem solving is discussed, and the opportunities such an approach affords for the future are considered.
...

References

SHOWING 1-10 OF 75 REFERENCES

Cooperative Plan Reasoning for Dialogue Systems

TLDR
Work in the TRAINS project consists of developing both formal models of the various tasks involved in collaborative dialogue and programs that implement these theories.

Integrating Natural Language Understanding and Plan Reasoning in the TRAINS-93 Conversation System

TLDR
The architecture of the system is described and particular attention is paid to the interactions between the language understanding and plan reasoning components, as well as how these two tasks constrain and inform each other in an integrated nl-based system.

Domain Plan Reasoning in TRAINS-90

TLDR
The domain plan reasoning aspects of the TRAINS-90 project is described, an integrated discourse-oriented system for natural language and planning research that provides an interface used by higher level routines to address issues that require reasoning about the world.

Knowledge Representation in the TRAINS System

TLDR
The goal of the paper is to describe how TRAINS-93, the latest prototype of TRAINS, deals with representational needs, representing lexical meaning, dealing with the problem of ambiguity, make use of information about context, and finding a connection between the content of the current utterance and the plan being jointly developed by system and user.

Language and Discourse in the TRAINS Project

TLDR
The general goals of the natural language aspects of the TRAINS project, including parsing, semantic interpretation and discourse modelling are described.

Arguing about Plans: Plan Representation and Reasoning for Mixed-initiative Planning

TLDR
A formal model of plans based on defeasible argument systems that allows for the ability to suggest aspects of a plan, accept such suggestions from other agents, criticize plans, revise them, etc., in addition to building plans is described.

The RHET System: A Sequence of Self-Guided Tutorials

TLDR
This report provides a sequence of tutorials each demonstrating a major feature of the RHET system, which consists of a set of separately defined specialized reasoning systems that are presented to the user within a single uniform framework.

The Discourse System Project

TLDR
This report describes an architecture for discourse systems that allows the integration of many different processing modules and describes an initial set of modules that have been implemented and tested using the architecture.

Conversational Events and Discourse State Change: A Preliminary Report

TLDR
It is argued that an action-based model of belief update is compatible with the proposals advanced in the literature on formal approach es to discourse interpretation, yet allows for a close integration between the processes tracking the attentional s tate and performing intention recognition.

Explicit Representation of Events, Actions and Plans for Assumption-Based Plan Reasoning

TLDR
This work begins with an event-based language based on the interval temporal logic that supports reasoning about action attempts and composite actions, and defines a representation for plans viewed as arguments that a certain course of action under certain explicit conditions will achieve certain goals.
...