Communicating contexts: A pragmatic approach to information exchange
In the ”Denk” project (”Dialogue handling and knowledge transfer”) we are implementing a multi-modal interface, combining graphics and language, that allows a user to interact with an object-oriented world. Using mouse and screen, the user has direct access to this world, but he can also converse about it with a so-called cooperative assistant, which happens to be an expert on the world in question. The cognitive state of this assistent is modelled in Type Theory. As has been pointed out in , , there are also good reasons to use Type Theory as a semantic representation formalism underlying natural language. If we combine these ideas, we can model cognitive states in Type Theory which change dynamically as a result of natural language utterances. In this way we can not only handle typical natural language phenomena, like anaphora resolution and the identification of presuppositions, but we can also mimic the different behaviours associated with questions and assertions. One of the key elements in our approach is that we model both the assistants' private knowledge about the world, and the common knowledge of user and assistant. These are represented by type theoretic contexts, where the latter is contained within the former. In a communication process, knowledge is transferred from the outer to the inner context, and the common knowledge grows steadily. Because knowledge states are not directly accessible for an outsider, the identity of objects can not be communicated directly between two agents, unless they use the fact that their knowledge is semantically attached to a world which they can both observe. This makes the communciation processes quite interesting, as will be illustrated for several simple cases.
Unable to display preview. Download preview PDF.
- 1.Alshawi, H. ‘Resolving Quasi Logical Forms', in: ‘Computational Linguistics’ 16, p133–144.Google Scholar
- 2.Barendregt, H.P. (1992) “Lambda calculi with types.” in: Handbook of Logic in Computer Science (Eds. S. Abramsky, D. Gabbay and T. Maibaum), Oxford University Press, Oxford.Google Scholar
- 3.Barendregt, H.P. and Hemerik K. (1990) ‘Types in lambda calculi and programming languages, in: European Symposium on Programming, ed. N. Jones, Lecture Notes in Computer Science 432, Springer, 1–36.Google Scholar
- 4.van Benthem, J. (1991): ‘Reflections on Epistemic Logic.’ Logique et Analyse, 133–134.Google Scholar
- 5.Borghuis, T. (1994): ‘Coming to Terms with Modal Logic’ Ph.D. thesis, Eindhoven University of Technology.Google Scholar
- 6.De Bruijn, N.G. (1980): ‘A Survey of the Project Automath.’ In: Seidin & Hindley (eds.): To H.B. Curry: Essays on Combinatory Logic, Lambda Calculus and Formalisms. Academic Press.Google Scholar
- 7.Curry, H.B. & R. Feys (1958): Combinatory Logic, Vol.1. Dordrecht: North Holland.Google Scholar
- 8.Ahn, R. & Kolb, H.P.: “Discourse Representation meets Constructive Mathematics.” In: Papers from the Second Symposium on Logic and Language. (L. Kálmán & L. Pólos, Eds.), Akadémia Kiadó, Budapest, 1990.Google Scholar
- 9.Mäenpää, P. & Ranta, A.: “An implementation of Intuitionistic Categorial Grammar” In: Papers from the Second Symposium on Logic and Language. (L. Káimán & L. Pólos, Eds.), Akadémia Kiadó, Budapest, 1990.Google Scholar
- 10.Kamp, H. (1981): ‘A Theory of Truth and Semantic Representation.’ In: J.A.G. Groenendijk et al. (eds.): Formal Methods in the Study of Natural Language. Amsterdam.Google Scholar
- 11.Mönnich, U. (1985): Untersuchungen zu einer konstruktiven Semantik für ein Fragment des Englischen. Habilitationsschrift, Tübingen University.Google Scholar