Dealing with Dependencies between Content Planning and Surface Realisation in a Pipeline Generation Architecture

The majority of existing language generation systems have a pipeline architecture which offers efficient sequential execution of modules, but does not allow decisions about text content to be revised in later stages. However, as exemplified in this paper, in some cases choosing appropriate content can depend on text length and formatting, which in a pipeline architecture are determined after content planning is completed. Unlike pipelines, interleaved and revision-based architectures can deal with such dependencies but tend to be more expensive computationally. Since our system needs to generate acceptable hypertext explanations reliably and quickly, the pipeline architecture was modified instead to allow additional content to be requested in later stages of the generation process if necessary.

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

[2]  Robert Rubinoff Integrating Text Planning and Linguistic Choice by Annotating Linguistic Structures , 1992, NLG.

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

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

[5]  Candace L. Sidner,et al.  Attention, Intentions, and the Structure of Discourse , 1986, CL.

[6]  R. Lathe Phd by thesis , 1988, Nature.

[7]  David Reinking,et al.  The Effects of Computer-Mediated Text on Measures of Reading Comprehension and Reading Behavior , 1985 .

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

[9]  Ian H. Beaumont,et al.  Workshop on adaptive hypertext and hypermedia , 1995, LINK.

[10]  Günter Neumann,et al.  POPEL-HOW: A Distributed Parallel Model for Incremental Natural Language Production with Feedback , 1989, IJCAI.

[11]  Y. Wilks,et al.  Artificial Believers: The Ascription of Belief , 1991 .

[12]  Ehud Reiter,et al.  Book Reviews: Building Natural Language Generation Systems , 2000, CL.

[13]  J. Davenport Editor , 1960 .

[14]  Maria Milosavljevic,et al.  Text Generation in a Dynamic Hypertext Environment , 2003 .

[15]  Johanna D. Moore Participating in explanatory dialogues , 1994 .

[16]  Cecile Paris User modelling in text generation , 1993 .

[17]  Kathleen McKeown,et al.  Discourse Strategies for Generating Natural-Language Text , 1985, Artif. Intell..

[18]  KambhampatiSubbarao,et al.  Planning as constraint satisfaction , 2001 .

[19]  Chris Mellish,et al.  An Architecture for Opportunistic Text Generation , 1998, INLG.

[20]  John A. Bateman Cécile L. Paris. User Modelling in Text Generation , 1996 .

[21]  Verzekeren Naar Sparen,et al.  Cambridge , 1969, Humphrey Burton: In My Own Time.

[22]  Cecile L. Paris User Modeling in Text Generation , 1993 .

[23]  Koenraad De Smedt,et al.  IPF: an incremental parallel formulator , 1990 .

[24]  Chris Mellish,et al.  Sources of Flexibility in Dynamic Hypertext Generation , 1996, INLG.

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

[26]  Jakob Nielsen,et al.  Designing Web Usability: The Practice of Simplicity , 1999 .