Generating under Global Constraints: the Case of Scripted Dialogue

Recently, the view of Natural Language Generation (nlg) as a Constraint Satisfaction Problem (csp) has seen something of a revival. The aim of this paper is to examine the issues that arise when nlg is viewed as a csp, and to introduce a novel application of constraint-based nlg, namely the Scripted Dialogue. Scripted Dialogue shares a number of crucial features with discourse, which make it possible to control the global properties of a computer-generated dialogue in the same way as those of a generated discourse. We pay particular attention to the use of soft constraints for enforcing global properties of text and dialogue. Because there has been little research into the formal properties of soft constraints in relation to generation, we start out with a theoretical exploration. We argue that, when multiple constraints are involved, it is important to define properly what is being optimised before proposing specific algorithms, and we argue that such definitions are often lacking in csp-based nlg. We show that it can be difficult (and sometimes even impossible) to guarantee satisfaction of global constraints by following local strategies. Based on these difficulties, we propose a novel approach to the generation of discourse and dialogue which combines csp solving with revision. Scripted Dialogue is used to illustrate this approach, which is compared with alternatives such as monitoring and estimation.

[1]  Richard Power,et al.  An integrated framework for text planning and pronominalisation , 2000, INLG.

[2]  Francesca Rossi,et al.  Soft Constraint Propagation and Solving in Constraint Handling Rules , 2004, Comput. Intell..

[3]  J. Nash.  THE BARGAINING PROBLEM , 1950, Classics in Game Theory.

[4]  Judith Masthoff,et al.  Group Modeling: Selecting a Sequence of Television Items to Suit a Group of Viewers , 2004, User Modeling and User-Adapted Interaction.

[5]  Richard Power.  Planning texts by constraint satisfaction , 2000, COLING.

[6]  Ehud Reiter.  Pipelines and Size Constraints , 2000, Computational Linguistics.

[7]  Eduard Hovy,et al.  Generating Natural Language Under Pragmatic Constraints , 1988 .

[8]  Richard Cox,et al.  Vicarious learning from dialogue and discourse -- A controlled comparison , 1999 .

[9]  Irene Langkilde-Geary.  An Exploratory Application of Constraint Optimization in Mozart to Probabilistic Natural Language Processing , 2004, CSLP.

[10]  Chris Mellish,et al.  Experiments Using Stochastic Search for Text Planning , 1998, INLG.

[11]  Thomas Rist,et al.  The automated design of believable dialogues for animated presentation teams , 2001 .

[12]  Michael Strube,et al.  Discrete Optimization as an Alternative to Sequential Processing in NLG , 2005, ENLG.

[13]  Daniel Marcu,et al.  Fast Decoding and Optimal Decoding for Machine Translation , 2001, ACL.

[14]  David L. Waltz,et al.  Understanding Line drawings of Scenes with Shadows , 1975 .

[15]  Amy Isard,et al.  Individuality and Alignment in Generated Dialogues , 2006, INLG.

[16]  Kees van Deemter,et al.  Constraint-based Natural Language Generation: A Survey , 2006 .

[17]  Kees van Deemter,et al.  Context modeling and the generation of spoken discourse , 1997, Speech Commun..

[18]  Johan Bos,et al.  A model of dia-logue moves and information state revision , 1999 .

[19]  Roger Evans,et al.  Empirically-based Control of Natural Language Generation , 2005, ACL.

[20]  Brigitte Krenn,et al.  RRL: A Rich Representation Language for the Description of Agent Behaviour in NECA , 2004, ArXiv.

[21]  Kentaro Inui,et al.  Text Revision: A Model and Its Implementation , 1992, NLG.

[22]  Kees van Deemter,et al.  Towards Automated Generation of Scripted Dialogue: Some Time-Honoured Strategies , 2003, ArXiv.

[23]  Kathleen McKeown,et al.  Empirically Designing and Evaluating a New Revision-Based Model for Summary Generation , 1996, Artif. Intell..

[24]  James C. Lester,et al.  Dynamically Imroving Explanations: A Revision-Based Approach to Explanation Generation , 1997, IJCAI.

[25]  Francesca Rossi,et al.  Semiring-based constraint solving and optimization , 1997 .

[26]  Henry S. Thompson,et al.  Towards A Computational Model Of Poetry Generation , 2000 .

[27]  Uwe Reyle,et al.  From Discourse to Logic - Introduction to Modeltheoretic Semantics of Natural Language, Formal Logic and Discourse Representation Theory , 1993, Studies in linguistics and philosophy.

[28]  J. Hayes,et al.  Writing Research and the Writer. , 1986 .

[29]  Arthur C. Graesser,et al.  Overhearing Dialogues and Monologues in Virtual Tutoring Sessions: Effects on Questioning and Vicarious Learning , 2000 .

[30]  Ehud Reiter,et al.  Has a Consensus NL Generation Architecture Appeared, and is it Psycholinguistically Plausible? , 1994, INLG.

[31]  David Traum,et al.  The Information State Approach to Dialogue Management , 2003 .

[32]  Azriel Rosenfeld,et al.  Scene Labeling by Relaxation Operations , 1976, IEEE Transactions on Systems, Man, and Cybernetics.

[33]  Didier Dubois,et al.  Possibility theory in constraint satisfaction problems: Handling priority, preference and uncertainty , 1996, Applied Intelligence.

[34]  Pascal Van Hentenryck.  Constraint satisfaction in logic programming , 1989, Logic programming.

[35]  Richard Power,et al.  Optimizing Referential Coherence in Text Generation , 2004, CL.

[36]  John A. Bateman,et al.  Sentence generation and systemic grammar: an introduction , 1997 .

[37]  A. Rubinstein.  Economics and Language , 1999 .

[38]  Kees van Deemter,et al.  Dialogue as Discourse: Controlling Global Properties of Scripted Dialogue , 2003, ArXiv.

[39]  Brigitte Krenn,et al.  Generation of multimodal dialogue for net environments , 2002 .

[40]  Paul Piwek,et al.  Natural Language Generation , 2004, Lecture Notes in Computer Science.

[41]  Roger Evans,et al.  A Framework for Stylistically Controlled Generation , 2004, INLG.