Summary of the paper

Title Context Fusion: The Role of Discourse Structure and Centering Theory
Authors Raffaella Bernardi, Manuel Kirschner and Zorana Ratkovic
Abstract Questions are not asked in isolation. Their context, viz. the precedinginteractions, might be of help to understand them and retrieve the correct answer. Previous research in Interactive Question Answering showedthat context fusion has a big potential to improve the performance of answerretrieval. In this paper, we study how much context, and what elements of it, should beconsidered to answer Follow-Up Questions (FU Qs). Following previous research, we exploit Logistic Regression Models to learnaspects of dialogue structure relevant to answering FU Qs. We enrich existing models based on shallow features with deep features,relying on the theory of discourse structure of (Chai and Jin, 2004), and on Centering Theory,respectively. Using models trained on realistic IQA data, we show which of the varioustheoretically motivated features hold up against empirical evidence.We also show that, while these deep features do not outperform the shallow ones on their own, an IQA system's answer correctnessincreases if the shallow and deep features are combined.
Language
Topics Dialogue, Discourse annotation, representation and processing
Full paper Context Fusion: The Role of Discourse Structure and Centering Theory
Bibtex @InProceedings{BERNARDI10.330,
  author = {Raffaella Bernardi, Manuel Kirschner and Zorana Ratkovic},
  title = {Context Fusion: The Role of Discourse Structure and Centering Theory},
  booktitle = {Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC'10)},
  year = {2010},
  month = {may},
  date = {19-21},
  address = {Valletta, Malta},
  editor = {Nicoletta Calzolari (Conference Chair), Khalid Choukri, Bente Maegaard, Joseph Mariani, Jan Odjik, Stelios Piperidis, Mike Rosner, Daniel Tapias},
  publisher = {European Language Resources Association (ELRA)},
  isbn = {2-9517408-6-7},
  language = {english}
 }
Powered by ELDA © 2010 ELDA/ELRA