UT boog University of Twente Home Page

SCHISMA

Dialogue systems

In the SCHISMA project we investigate different possible architectures and techniques for building natural language dialogue systems. It serves as a motivation and testbed for various approaches and theories. SCHISMA is both the name of the project and the name of our prototype system: a dialogue system for theatre information and booking. The acronym is derived from the Dutch SCHouwburg Informatie Systeem. SCHISMA started as a joint research project of PTT Research and the Parlevink group of the University of Twente. The Twentse Schouwburg provides us with the necessary information to fill our database. The SCHISMA project has lead to publications, a prototype, a corpus and automatically generated grammars, work on spoken dialogue and various other activities related to dialogue systems, such as a virtual theatre and graduation work by students.

Motivation

We believe that natural language dialogue systems have several advantages over command-languages or graphical user interfaces. Natural language is natural to the user; no need to learn an austere command language or master the graphical interface controls. When using speech, it is even more natural. Moreover, speech technology opens the possibility to use the telephone for automatic information services. The use of dialogue, i.e. an interactive exchange of meaningful utterances, makes it possible to correct misunderstanding and recognition errors. In dialogue, circumstances and previous utterances play a role: the dialogue context. The context can be used, not only for disambiguation, but also to make information access more easy to use and more effective. Users can continue to ask queries about the previous topic. The system can remember user characteristics it has asked before.

The disadvantages of natural language interfaces have to be carefully dealt with. For instance, natural language interfaces do not naturally indicate the limits of the capacities and vocabulary of the system. Users will often be irritated by a missing functionality or unknown word. Such problems can however be overcome by a careful design of the systems `prompts' and thus manipulating the users perception of the system.

The application chosen for our prototype is a theatre information and booking service. The reason we choose theatre information, is that the domain is richer then the usual train- or airline information services. We could add all kinds of general knowledge about Shakespeare, Balanchine and so on. However, when restricted to the theatre schedule - what performance plays at what day - it can be constructed just as easily as the public transport information systems.

The second interesting feature of our application, is that it involves reservations. The user may reserve tickets for a particular performance. From the perspective of the system, a reservation requires a unique performance, a number of tickets and information on discount, preferred seats, as well as the name and identification number of the user. The user however, would first expect detailed information on the price and seats before giving his consent. We like to look at a reservation transaction as a promise of the user to pick up the tickets half an hour in advance and pay the price, given that the information the system provided was correct. In future we hope to extend the prototype into a system that is able to give advice or help the user make up his or her mind.

Corpus and Grammar

Building a dialogue system requires a corpus. The corpus gives insights into the behaviour of users and the typical expressions used in a particular domain. With a corpus the coverage of the lexicon and grammar can be evaluated and improved. In a Wizard-of-Oz experiment we collected a large sample of dialogues. The corpus is in Dutch and has been annotated and described. ( to the Corpus (zipped)).

Hugo ter Doest has generated a probabilistic unification grammar from the corpus. The corpus was tagged with syntactic categories and superficial structure. Unification constraints and probabilities were derived. The grammar was then tested on `seen' and `unseen' data from the same domain using a probabilistic left-corner parser for PATRII unification grammars. For more information see here or the list of publications (Ter Doest, 1998).

SCHISMA Prototype (THIS v 1.0)

Try the SCHISMA Testapplet

Our current version of the SCHISMA system is called THIS v1.0 (THeatre Information System). It was build by Danny lie, Bas Leijdekkers and Eric Barten. The parser is based on a rewrite approach. The system has a simple agent architecture. Here is short description. For a more extensive description, see the list of publications (Lie et al 1997).

Virtual Theatre

In the SCHISMA project we investigate natural language dialogue systems in general. Dialogue is a natural way of accessing information. However, natural interaction is not necessarily restricted to speech or keyboard input. We are interested in other forms of communication as well. One such possibility, is the spatial metaphor: the user can make a virtual walk through the theatre. Inside the theatre information can be found at various locations. Information may be presented in the form of brochures, posters or in the form of an artificial agent Karin that is able to answer questions about performances and make reservations.

A virtual browser based on the VRML technology was developed by Richard Selles and Eduard Selles (Hogeschool Enschede, HIO). It contains the floor-plan of the Muziekcentrum building and various types of information related to theatre and music. At the moment, we are connecting the Karin agent interface to our SCHISMA prototype.

More information on the Virtual Theatre is available in Dutch. The VRML browser requires specific software and hardware. Here is a description of the system requirements and how to download them. Here is a list of screenshots from the Theatre.

Spoken Dialogue Systems

In several student projects we acquired experience in building telephone-based dialogue systems for transactions, especially ticket reservation. Our experiences with the development toolkit, Speechmania, and some of our recommendations and conclusions can be found at Dialogue Systems for Reservation.
  • During his external project at the Telecats, Mathieu van den Berk has build a reservation system for a movie theatre. You can still try it at : 053-4889961.
  • As part of an external project with Coopers and Lybrand, Karianne Braat and Harold Jurgen Oortwijn have build an automatic flight reservation system for a small Irish airline company. You might try it at : 053 - 4894868

Other Activities

  • Joris Hulstijn developed a small scale Prolog version of the SCHISMA prototype, called SCHISMINI. It is designed as a specification of the dialogue actions of the dialogue manager, and can be applied in teaching.

Other Theatre Information Systems...

People

The following people are associated with the SCHISMA project.

Previous Contributors

  • Toine Andernach (now at Leiden University, speech act classification, machine learning)
  • Stan van de Burgt (PTT Research)
  • Hans Kloosterman (PTT Research)

Graduation work

SCHISMA motivates a lot of student research. Here is a list of recent graduation projects on dialogue systems. For more on graduation projects, see our student information page.
  • Gies Bouwman (Februari 1998) Dialogue Strategies in Interface Design (external project at Philips Research Laboratories, Aachen)
  • Thijs Westerveld (June 1997) Contextgevoelige Spelling Correctie in SCHISMA
  • Sander van der Berg (June 1997) Supervised Classification Algoritmen en hun gebruik voor het classificeren van uitingen in een natuurlijke taal dialoogsysteem
  • Etto Salomons (September 1996) The Use of Kohonen Networks in Finding Classes of Dialogue Utterances
  • René Steetskamp (July 1996) Semantics in the Schisma Domain
  • René Bos (1995) Modelling Dialogues with Finite State Automata

Publications

1998 1997
  • D.H. Lie, J. Hulstijn H. ter Doest & A. Nijholt. Language analysis in Schisma. In: International Workshop on Parsing Technologies 1997 (IWPT '97), MIT, Boston, USA, 1997, 243-244.
    postscript version
  • T. Andernach, M. Poel and E. Salomons. Finding Classes of Dialogue Utterances with Kohonen Networks. In: Proceedings NLP workshop of the European Conference on Machine Learning (ECML), Prague, April 1997.
  • J. Hulstijn, R. Steetskamp, H. ter Doest, S.P. van de Burgt & A. Nijholt. Dialogues in a theatre information and booking system. In: Memorias. V Symposio Internacional de Comunicacion Social. Santiago de Cuba, Centro de Linguistica Aplicada, 1997, 87-99.
    postscript version
  • A. Nijholt. Human-computer interaction and computer-telephony integration. In: FIE'97, Universidad de Oriente, Facultad de Ingenieria Electrica, Santiago de Cuba, 1997. To appear.
  • A. Nijholt and T. Andernach. The Robust Analysis of User Utterances in a Mixed-initiative Dialogue System. Fifth International Symposium on Social Communication, Santiago de Cuba, 1997, p.31.
  • R. Leenes & J. Schaake. OLE2000: A study in questions and demand. Presented at the IFIP WG 8.5
  • J. Hulstijn. Structured information states - raising and resolving issues. In: Proceedings of Mundial'97, Formal Semantics and Pragmatics of Dialogue. G. Jäger and A. Benz (eds.), University of Munich, 1997, to appear.
    Also available as CTIT technical report 97-18, ISSN 1381 - 3625, 20 pp.
1996
  • G.Veldhuijzen van Zanten, S. Luperfoy and A. Nijholt (eds.). Dialogue Management in Natural Language Processing. Proceedings Twente Workshop on Language Technology 11 (TWLT11), University of Twente, June 1996
  • J. Hulstijn, R. Steetskamp, H. ter Doest, S.P. van de Burgt & A. Nijholt. Topics in SCHISMA Dialogues. In: Proceedings Twente Workshop on Language Technology 11 (TWLT11), University of Twente, June 1996
  • H. ter Doest, M. Moll, R. Bos & A. Nijholt. Language Engineering in Dialogue Systems. In: Proceedings Computers in Engineering Symposium. Session on Natural Language in Human-Computer Interfaces. Houston, Texas, February 1996
  • G.F. van der Hoeven and A. Nijholt. Casting actors in a context-free play. Publicationes Mathematicae, Debrecen, Tom.48, Fasc. 1-2, Automata and Formal Languages, Supplement (1996), 291-308.
  • S.P. van de Burgt, H. Kloosterman, T. Andernach, R. Bos & A. Nijholt. Building dialogue systems that sell. Proceedings International Conference on Natural Language Processing and Industrial Applications. June 1996, Moncton, New Brunswick, Canada
1995
  • T. Andernach, S.P. van de Burgt & G.F. van der Hoeven (eds.). Corpus-Based Dialogue Modelling. Proceedings Twente Workshop on Language Technology 9 (TWLT9), University of Twente, June 1995, 124 pp.
  • T. Andernach. Predicting and interpreting speech acts in a theatre information and booking system. In: Proc. Twente Workshop on Language Technology 9 (TWLT9), T. Andernach, S.P. van de Burgt & G.F. van der Hoeven (eds.), University of Twente, Enschede, the Netherlands, June 9, 1995, 107-115.
  • G.F. v.d. Hoeven, J.A. Andernach, S.P. v.d. Burgt, G-J.M. Kruijff, A. Nijholt, J. Schaake, and F.M.G. de Jong. SCHISMA: A Natural Language Accessible Theatre Information and Booking System. Proceedings First International Workshop on Applications of Natural Language to Data Bases, Versailles, France, June 1995, 271-285.
  • R. op den Akker, H. ter Doest, M. Moll & A. Nijholt. Parsing in dialogue systems using typed feature structures. In: Proceedings 4th International Workshop on Parsing Technologies (IWPT'95), Prague, Karlovy Vary, 1995, 10-11.
  • T. Andernach, H. ter Doest, R. op den Akker, G. van der Hoeven, S.P. van de Burgt, J. Schaake & A. Nijholt. Language Analysis for Dialogue Management in a Theatre Information & Booking System. Language Engineering, AI 95, 15th International Conference, Montpellier, June 1995, 351-362.
  • G.-J.M. Kruijff & J. Schaake. Discerning relevant information in discourses using TFA. In: Proceedings Recent Advances in Natural Language Processing, Velingrad, Bulgaria, September 1995, 214-220.
  • J. Schaake & G.-J.M. Kruijff. Information states based analysis of dialogues. In: Proceedings Recent Advances in Natural Language Processing, Velingrad, Bulgaria, September 1995, 271-277.
  • R. Bos. Modelling dialogues with finite automata in SCHISMA. Report R&D-SV-95-144. KPN Research, Leidschendam, March 1995, 49 p.
  • E. Komen. Evaluation of Natural Language for the Schisma domain. Memoranda Informatica 95-14, March 1995, 31 pp.
1994
  • G.F. van der Hoeven, S.P. van de Burgt, T. Andernach, J. Schaake, G.-J.M. Kruijff, A. Nijholt & F.M.G. de Jong. SCHISMA: A natural language accessible theatre information and booking system. In: Speech and Language Engineering. Proceedings Twente Workshop on Language Technology 8 (TWLT8), L. Boves & A. Nijholt (eds.), University of Twente, The Netherlands, 1994, 137-149, ISNN 0929-0672.
  • G. Veldhuizen van Zanten & R. op den Akker. Developing natural language interfaces: A test case. In: Speech and Language Engineering. Proceedings Twente Workshop on Language Technology 8 (TWLT8), L. Boves & A. Nijholt (eds.), University of Twente, The Netherlands, 1994, 121-135, ISSN 0929-0672.
  • T. Andernach & M. van Steenbergen. Domain and dialogue knowledge in a natural language information system. Memoranda Informatica 94-05, January 1994, 55 pp., ISSN 0924-3755.
  • R. op den Akker, T. Andernach, S. van de Burgt, G. van der Hoeven, F.M.G. de Jong, A. Nijholt & M. van Steenbergen. Natuurlijke taal interfaces voor dialoogsystemen. Memoranda Informatica 94-04, January 1994, 41 pp., ISSN 0924-3755.

    Last modified $Date: 2003/07/21 07:32:13 $ by Joris Hulstijn