A Computational Architecture for Conversation

  • Eric Horvitz
  • Tim Paek
Part of the CISM International Centre for Mechanical Sciences book series (CISM, volume 407)


We describe representation, inference strategies, and control procedures employed in an automated conversation system named the Bayesian Receptionist. The prototype is focused on the domain of dialog about goals typically handled by receptionists at the front desks of buildings on the Microsoft corporate campus. The system employs a set of Bayesian user models to interpret the goals of speakers given evidence gleaned from a natural language parse of their utterances. Beyond linguistic features, the domain models take into consideration contextual evidence, including visual findings. We discuss key principles of conversational actions under uncertainty and the overall architecture of the system, highlighting the use of a hierarchy of Bayesian models at different levels of detail, the use of value of information to control question asking, and application of expected utility to control progression and backtracking in conversation.


Bayesian Network Bayesian Model Natural Language Processing Computational Architecture Front Desk 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Albrecht, D.W., Zukerman, I., Nicholson, A.E., Bud, A. (1997). Towards a Bayesian model for keyhole plan recognition in large domains. In Jameson, A, Paris, C., and Tasso, C., eds., Proceedings of the Sixth International Conference on User Modeling. New York: Springer-Verlag. 365–376.Google Scholar
  2. Allen, J.F., and Perrault, C. (1980). Analyzing intention in utterances. Artificial Intelligence 15:143–178.CrossRefGoogle Scholar
  3. Clark, H.H. (1983). Making sense of nonce sense. In G.B. Flores d’Arcais and R. Jarvella, eds., The Process of Language Understanding. New York: Wiley. 297–331.Google Scholar
  4. Clark, H.H. (1996). Using Language. Cambridge University Press.CrossRefGoogle Scholar
  5. Cohen, P.R., and Levesque, H.J. (1994). Preliminaries to a collaborative model of dialogue. Speech Communication 15:265–274.CrossRefGoogle Scholar
  6. Conati, C., Gertner, A.S., VanLehn, K., Druzdzel, M.J. (1997). Online student modeling for coached problem solving using Bayesian networks. In Jameson, A, Paris, C., and Tasso, C., eds., Proceedings of the Sixth International Conference on User Modeling. New York: Springer-Verlag. 231–242.Google Scholar
  7. Gorry, G.A. and Barnett, G.O. (1968). Experience with a model of sequential diagnosis. Computers and Biomedical Research 1:490–507.CrossRefGoogle Scholar
  8. Heckerman, D.E., Horvitz, E., and Nathwani, B.N. (1992). Toward normative expert systems: Part I. The Pathfinder project. Methods of Information in Medicine 31:90–105.Google Scholar
  9. Heckerman, D., and Horvitz, E. (1998). Inferring informational goals from free-text queries: A Bayesian approach, Fourteenth Conference on Uncertainty in Artificial Intelligence, San Francisco: Morgan Kaufmann Publishers, 230–237. Scholar
  10. Heidorn, G.(1999) Intelligent writing assistance. In Dale, R., Moisl, H., and Somers, H. eds., A Handbook of Natural Language Processing Techniques. Marcel Dekker.Google Scholar
  11. Horvitz, E. Heckerman, D.E., Ng, K. and Nathwani, B.N. (1989). Heuristic abstraction in the decision-theoretic Pathfinder system, Proceedings of the Thirteenth Symposium on Computer Applications in Medical Care. IEEE Computer Society Press. 178–182.Google Scholar
  12. Horvitz, E., Breese, J., and Henrion, M. (1988). Decision theory in expert systems and artificial intelligence. International Journal of Approximate Reasoning, Special Issue on Uncertainty in Artificial Intelligence 2:247–302. Scholar
  13. Horvitz, E., Barry, M. (1995). Display of information for time-critical decision making. In Besnard, P., and Hanks, S.,. eds., Proceedings of the Eleventh Conference on Uncertainty in Artificial Intelligence. San Francisco: Morgan Kaufmann. 296–305. Scholar
  14. Horvitz, E., Breese, J.S., Heckerman, D., Hovel, D., Rommeise, K. (1998). The Lumiere Project: Bayesian user modeling for inferring the goals and needs of software users, Fourteenth Conference on Uncertainty in Artificial Intelligence. San Francisco: Morgan Kaufmann Publishers. 256–265. Scholar
  15. Horvitz, E. (1999). Principles of mixed-initiative user interfaces, In Proceedings of Computer-Human Interaction ’99, Association for Computing Machinery Press.Google Scholar
  16. Jameson, A., Schafer, R., Simons, J., and Weis, T. (1995). Adaptive provision of evaluation-oriented information: Tasks and techniques. Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence. 1886–1895.Google Scholar
  17. Jameson, A. (1996) Numerical uncertainty management in user and student modeling: An overview of systems and issues. User Modeling and User-Adapted Interaction 5:193–251.CrossRefGoogle Scholar
  18. Jennings, N.R., and Mamdani, E.H. (1992). Using joint responsibility to coordinate collaborative problem solving in dynamic environments. Proceedings of the Tenth National Conference on Artificial Intelligence. Menlo Park: AAAI Press. 269–275.Google Scholar
  19. Levinson, S.C. (1992). Activity types and language. In P. Drew and J. Heritage, eds., Talk at Work. Cambridge University Press. 66–100.Google Scholar
  20. Nunberg, G. (1979). The non-uniqueness of semantic solutions: Polysemy. Linguistics and Philosophy 3:143–184.CrossRefGoogle Scholar
  21. Paek, T., and Horvitz, E. (1999), A layered representation of uncertainty for managing dialogue. Manuscript submitted for publication.Google Scholar

Copyright information

© Springer Science+Business Media New York 1999

Authors and Affiliations

  • Eric Horvitz
    • 1
  • Tim Paek
    • 2
  1. 1.Decision Theory and Adaptive SystemsMicrosoft ResearchUSA
  2. 2.Department of PsychologyStanford UniversityUSA

Personalised recommendations