A bayesian approach to intention-based response generation

The statistical approach to natural language generation of overgeneration-andranking suffers from expensive over generation. This article reports the findings of response classification experiment in the new approach of intention-based classification-andranking. Possible responses are deliberately chosen from a dialogue corpus rather than wholly generated, so the approach allows short ungrammatical utterances as long as they satisfy the intended meaning of the input utterance. We hypothesize that a response is relevant when it satisfies the intention of the preceding utterance, therefore this approach highly depends on intentions, rather than syntactic characterization of input utterance. The response classification experiment is tested on a mixed-initiative, transaction dialogue corpus in the theater domain. This article reports a promising start of 73% accuracy in prediction of response classes in a classification experiment with application of Bayesian networks.

[1]  David R. Traum,et al.  CONVERSATION ACTS IN TASK‐ORIENTED SPOKEN DIALOGUE , 1992, Comput. Intell..

[2]  Chris Mellish,et al.  Conversation in the museum: experiments in dynamic hypermedia with the intelligent labelling explorer , 1998, New Rev. Hypermedia Multim..

[3]  Joris Hulstijn,et al.  Dialogue Models for Inquiry and Transaction , 2000 .

[4]  Finn V. Jensen,et al.  Bayesian Networks and Decision Graphs , 2001, Statistics for Engineering and Information Science.

[5]  Gwyneth Doherty-Sneddon,et al.  The Reliability of a Dialogue Structure Coding Scheme , 1997, CL.

[6]  Srinivas Bangalore,et al.  Learning the Structure of Task-Driven Human-Human Dialogs , 2006, ACL.

[7]  Amanda Stent,et al.  A conversation acts model for generating spoken dialogue contributions , 2002, Comput. Speech Lang..

[8]  Rieks op den Akker,et al.  Dialogue act recognition under uncertainty using Bayesian networks , 2007, Natural Language Engineering.

[9]  Michael Strube,et al.  Classification-Based Generation Using TAG , 2004, INLG.

[10]  Kevin Knight,et al.  Generation that Exploits Corpus-Based Statistical Knowledge , 1998, ACL.

[11]  James F. Allen,et al.  Draft of DAMSL Dialog Act Markup in Several Layers , 2007 .

[12]  Harry Bunt,et al.  Dialogue pragmatics and context specification , 2000, Abduction, Belief and Context in Dialogue.

[13]  Marilyn A. Walker,et al.  An Application of Reinforcement Learning to Dialogue Strategy Selection in a Spoken Dialogue System for Email , 2000, J. Artif. Intell. Res..

[14]  J. Austin.  How to do things with words , 1962 .

[15]  D. Sperber,et al.  Relevance: Communication and Cognition , 1997 .

[16]  Benoit Lavoie,et al.  A Fast and Portable Realizer for Text Generation Systems , 1997, ANLP.

[17]  Adwait Ratnaparkhi,et al.  Trainable approaches to surface natural language generation and their application to conversational dialog systems , 2002, Comput. Speech Lang..

[18]  Anja Belz,et al.  Statistical Generation: Three Methods Compared and Evaluated , 2005, ENLG.

[19]  Sandra Carberry,et al.  Plan Recognition in Natural Language Dialogue , 1990 .

[20]  Marilyn A. Walker,et al.  Mixed Initiative in Dialogue: An Investigation into Discourse Segmentation , 1990, ACL.

[21]  Joris Hulstijn,et al.  Utterance generation for transaction dialogues , 1998, ICSLP.

[22]  Irene Langkilde.  Forest-Based Statistical Sentence Generation , 2000, ANLP.

[23]  Norbert Reithinger,et al.  Learning dialogue structures from a corpus , 1997, EUROSPEECH.

[24]  Chris Mellish,et al.  Using Classification to Generate Text , 1992, ACL.

[25]  Anton Nijholt,et al.  SCHISMA: A Natural Language Accessible Theatre Information and Booking System , 1995, NLDB.

[26]  E. Schegloff,et al.  Opening up Closings , 1973 .

[27]  Srinivas Bangalore,et al.  Exploiting a Probabilistic Hierarchical Model for Generation , 2000, COLING.

[28]  David Traum,et al.  Speech Acts for Dialogue Agents , 1999 .