A SVM Cascade for Agreement/Disagreement Classification

This article describes a method for classifying dialogue ut terances and detecting the interlocutor's agreement or disagreement. This labellingcan help improve dialogue manage- ment by providing additional information on the utterance' s content without deep parsing. The proposed technique improves upon state of the art approache s by using a Support Vector Ma- chine cascade. A combination of three binary support vector machines in a c ascade is employed to filter out utterances that are easy to classify, thus reducing the noise in the learning of labels for more ambiguous utterances. The approach achieves highe r accuracy (by 2.47%) than the state of the art while using a simpler approach which relies o on shallow local features of the utterances. RESUME. Dans cet article, nous decrivons une methode de classificati on d'utterances destinee a la detection d'accord/desaccord dans le dialogue homme-m achine. L'etiquetage du dialogue peut etre utilise par le dialogue manager sans avoir a effect uer de parse complexe. Nous propo- sons une technique de classification a base d'une hierarchiede classificateurs Support Vector Machines. Une combinaison de trois classificateurs binaire s est utilisee pour filtrer les classes pour lesquelles le corpus contient beaucoup d'informationet se concentrer sur les classes plus ambigues. Cet article presente une analyse detaillee des tr aits caracteristiques de classifica- tion et propose une amelioration de 2.47% sur l'etat de l'arttout en utilisant un modele de classification plus performant.

[1]  Chris Reed,et al.  Argumentation Machines, New Frontiers in Argument and Computation , 2004, Argumentation Machines.

[2]  Julia Hirschberg,et al.  Identifying Agreement and Disagreement in Conversational Speech: Use of Bayesian Networks to Model Pragmatic Dependencies , 2004, ACL.

[3]  Andreas Stolcke,et al.  Dialogue act modeling for automatic tagging and recognition of conversational speech , 2000, CL.

[4]  Rosalind W. Picard,et al.  Dialog Act Classification from Prosodic Features Using Support Vector Machines , 2002 .

[5]  Toine Andernach A Machine Learning Approach to the Classification of Dialogue Utterances , 1996, ArXiv.

[6]  Justine Cassell,et al.  Negotiated Collusion: Modeling Social Language and its Relationship Effects in Intelligent Agents , 2003, User Modeling and User-Adapted Interaction.

[7]  Andreas Stolcke,et al.  The ICSI Meeting Corpus , 2003, 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)..

[8]  Valeria Carofiglio,et al.  Portia: A User-Adapted Persuasion System in the Healthy-Eating Domain , 2007, IEEE Intelligent Systems.

[9]  Suresh Manandhar,et al.  Argumentative Human Computer Dialogue for Automated Persuasion , 2008, SIGDIAL Workshop.

[10]  James A. Moore,et al.  Dialogue-Games: Metacommunication Structures for Natural Language Interaction , 1977, Cogn. Sci..

[11]  Dan Jurafsky,et al.  Dialog Act Modeling for Conversational Speech , 1998 .

[12]  S. Jackson,et al.  Structure of conversational argument: Pragmatic bases for the enthymeme , 1980 .

[13]  Alessandro Moschitti,et al.  Annotating Spoken Dialogs: From Speech Segments to Dialog Acts and Frame Semantics , 2009 .

[14]  H. Kotthoff,et al.  Disagreement and concession in disputes: On the context sensitivity of preference structures , 1993, Language in Society.

[15]  K. Bennett,et al.  A support vector machine approach to decision trees , 1998, 1998 IEEE International Joint Conference on Neural Networks Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36227).

[16]  Jennifer G. Dy,et al.  A hierarchical method for multi-class support vector machines , 2004, ICML.

[17]  Mari Ostendorf,et al.  Detection Of Agreement vs. Disagreement In Meetings: Training With Unlabeled Data , 2003, NAACL.

[18]  Floriana Grasso,et al.  The Persuasion Machine , 2004, Argumentation Machines.

[19]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[20]  Vladimir Vapnik,et al.  The Nature of Statistical Learning , 1995 .

[21]  A. Stolcke,et al.  Dialog act modelling for conversational speech , 1998 .