IM2.MDM Public Documents: Reports

  1. Georgescul M., Popescu-Belis A. & Starlander M. (2002) - Outils informatiques de manipulation de la vidéo et du son : une introduction. Project Report IM2.MDM-01, August 2002, 7 pages.
  2. Pallotta, V. (2003) - Computational Dialogue Models. Project Report IM2.MDM-02, March 2003, 81 pages.
  3. Clark A. (2003) - Machine Learning Approaches to Shallow Parsing: A Literature Review. Project Report IM2.MDM-03, March 2003, 9 pages.
  4. Georgescul M. (2003) - SOAP Webservice for querying the IM2.MDM dialogue database: Java package documentation. Project Report IM2.MDM-04, April 2003, 51 pages.
  5. Popescu-Belis A. (2003) - Resources, Tools and Projects for Multimodal Dialogue Understanding and Management: a Web-based Review. Project Report IM2.MDM-05, June 2003, 17 pages.
  6. Popescu-Belis A., Clark A., Georgescul M., Starlander M. & Zufferey S. (2003)A Thematic Bibliography on Dialogue Processing. Project Report IM2.MDM-06, June 2003, 27 pages.
  7. Popescu-Belis A. (2003) - Shallow Dialogue Analysis: Definition, Annotation, Visualisation. Project Report IM2.MDM-07, July 2003, 39 pages.
  8. Pallotta V. & Ghorbel H. (2003) - Argumentative Segmentation and Annotation Guidelines. Project Report IM2.MDM-08, June 2003, 11 pages.
  9. Popescu-Belis A. (2003) - Dialogue act tagsets for meeting understanding: an abstraction based on the DAMSL, Switchboard and ICSI-MR tagsets. Project Report IM2.MDM-09, v1.2, December 2004, 51 pages.
  10. Georgescul M. & Popescu-Belis A. (forthcoming) - Database and TQB Demonstrator: Installation Manual. Project Report IM2.MDM-10, (forthcoming).
  11. Lisowska, A. (2003) - Multimodal Interface Design for the Multimodal Meeting Domain: Preliminary Indications from a Query Analysis Study. Project Report IM2.MDM-11, November 2003, 30 pages.
  12. Clark A. & Popescu-Belis A. (2004) - Multilevel Dialogue Acts and Feature Selection. Project Report IM2.MDM-12, WP.STAT Deliverable, March 2004, 19 pages.
  13. Popescu-Belis A. & Palacio E. (forthcoming) - Data formatting and conversion procedures: summary of results. Project Report IM2.MDM-13, forthcoming.
  14. Lalanne D., Lisowska A., Bruno E., Flynn M., Georgescul M., Guillemot M., Janvier B., Marchand-Maillet S., Melichar M., Moenne-Loccoz N., Popescu-Belis A., Rajman M., Rigamonti M., von Rotz D. & Wellner P. (2005) - The IM2 Multimodal Meeting Browser Family. Joint IM2 Technical Report, March 2005, 17 p.
  15. Lüthi E. (2005) - Approaches to Topic Annotation of Discourse in Meeting Contexts. Project Report IM2.MDM-15, November 2005, 20 pages.

  • Report IM2.MDM-01, August 2002
    • Outils informatiques de manipulation de la vidéo et du son : une introduction
    • Maria Georgescul, Andrei Popescu-Belis, Marianne Starlander (ISSCO/TIM/ETI/UniGe)
    • Abstract: Report on tools for video/sound manipulation and conversion between different formats (in French, 7 pages)
    • PDF

  • Report IM2.MDM-02, March 2003
    • Computational Dialogue Models
    • Vincenzo Pallotta (LITH/IC/EPFL)
    • Abstract: This report reviews the theoretical foundations of Dialogue Management Systems, and in particular the problems involved in developing systems that can participate in natural dialogue with humans (MDM.EPFL.LITH.Y1.M6 Deliverable, 81 pages, also published as EACL '2003 Tutorial)
    • PDF

  • Report IM2.MDM-03, March 2003
    • Machine Learning Approaches to Shallow Parsing: A Literature Review
    • Alexander Clark (ISSCO/TIM/ETI/UniGe)
    • Abstract: This document reviews the literature on shallow discourse parsing, in particular the use of machine learning techniques. This is deliverable Y1.M6 of the IM2.MDM Discourse Parsing White Paper (9 pages)
    • PDF

  • Report IM2.MDM-04, April 2003
    • SOAP Webservice for querying the IM2.MDM dialogue database: Java package documentation
    • Maria Georgescul (ISSCO/TIM/ETI/UniGe)
    • Abstract: This is a printable version of the Java documentation of the SOAP Webservice that handles external queries to the IM2.MDM dialogue database (51 pages). For more information, please contact the author, Maria.Georgescul@eti.unige.ch.
    • PDF

  • Report IM2.MDM-05, June 2003
    • Resources, Tools and Projects for Multimodal Dialogue Understanding and Management: a Web-based Review
    • Andrei Popescu-Belis (ISSCO/TIM/ETI/UniGe)
    • Abstract: This report reviews recent initiatives and resources in the field (17 pages)
    • HTML | PDF

  • Report IM2.MDM-06, June 2003
    • A Thematic Bibliography on Dialogue Processing
    • Andrei Popescu-Belis, Alexander Clark, Maria Georgescul, Marianne Starlander, Sandrine Zufferey (ISSCO/TIM/ETI/UniGe)
    • Abstract: This report gathers important references in the field, annotated using keywords (24 pages and EndNote database file). The EndNote and BibTeX files are available upon request to project members.
    • PDF | HTML

  • Report IM2.MDM-07, July 2003
    • Shallow Dialogue Analysis: Definition, Annotation, Visualisation
    • Andrei Popescu-Belis (ISSCO/TIM/ETI/UniGe)
    • Abstract: This report provides the formal definition of the IM2.MDM shallow dialogue analysis protocol, guidelines for annotation (formats, tools) and formatting instructions (XSLT stylesheets + SMIL output) for HTML visualisation and database input.
    • Version 1.0 - PDF

  • Report IM2.MDM-08, June 2003
    • Argumentative Segmentation and Annotation Guidelines
    • Vincenzo Pallotta, Hatem Ghorbel (LITH/IC/EPFL)
    • Abstract: Definition of an Initial Argumentative Grammar based on analysis of six ICSI meeting dialogues. (11 pages)
    • PDF

  • Report IM2.MDM-09, September 2003
    • Dialogue act tagsets for meeting understanding: an abstraction based on the DAMSL, Switchboard and ICSI-MR tagsets
    • Andrei Popescu-Belis (ISSCO/TIM/ETI/UniGe)
    • Abstract: This report analyzes the DAMSL, SWBD and ICSI-MR tagsets for dialogue act annotation, then proposes two tagsets: MALTUS and Primula. MALTUS is an abstraction over the ICSI-MR set that attempts to reduce the number of possible labels. Primula is a multidimensional tagset based on different pragmatic theories of the posible functions of utterances in dialogue. Several conversion tables are provided between the various tagsets. In particular ICSI-MR data is converted to MALTUS, and its consistencs is checked in the process. (45 pages)
    • V1.2, December 2004 - PDF

  • Report IM2.MDM-10, October 2003 (forthcoming)
    • IM2.MDM Demonstrator: Installation Manual
    • Maria Georgescul, Andrei Popescu-Belis (ISSCO/TIM/ETI/UniGe); David Portabella, Martin Rajman (LIA/IC/EPFL)
    • Abstract: This report outlines the main software components and prerequisites for the various elements of the IM2.MDM demonstrator: dialogue database, web service, multimodal interface.

  • Report IM2.MDM-11, November 2003
    • Multimodal Interface Design for the Multimodal Meeting Domain: Preliminary Indications from a Query Analysis Study
    • Agnes Lisowska (ISSCO/TIM/ETI/UniGe)
    • Abstract: The design of a multimodal interface for retrieving the content of multimodal meeting recordings from a database is a complicated process that requires careful consideration of the needs of real users of such an interface in real use situations. This report describes the results of a study in which potential queries to such an interface were elicited from real users, with the aim of developing a basic set of user requirements for interface design. Results and indications extrapolated from an analysis of the queries are discussed, and areas where further research is required are highlighted. Finally, the report discusses future directions to pursue in order to continue and refine the design process, and how the data set that was collected can be used for the design and evaluation of components other than the interface itself. (30 pages)
    • Report (PDF) | List of all queries (plain text)

  • Report IM2.MDM-12 / Discourse Parsing WP Deliverable, March 2004
    • Multilevel Dialogue Acts and Feature Selection
    • Alexander Clark and Andrei Popescu-Belis (ISSCO/TIM/ETI/UniGe)
    • Abstract: This report discusses the use of multi-layered tagsets for dialogue acts, in the context of dialogue understanding for multi-party meeting recording and retrieval applications. We discuss some desiderata for such tagsets and critically examine some previous proposals. We then dene MALTUS, a new tagset based on the ICSI-MR and Switch- board tagsets, which satises these requirements. We present some experiments using MALTUS which attempt to compare the merits of integrated versus multi-level classiers for the detection of dialogue acts, and discuss some simple contextual features that are useful in this task. (19 pages)
    • PDF

  • Report IM2.MDM-13 (forthcoming).
    • Data formatting and conversion procedures: summary of results
    • Andrei Popescu-Belis and Emmanuel Palacio
    • Abstract: This report explains how various dialogue resources can be converted from various input formats to the SDA.XML structure, then to HTML and tabular format. The report also outlines the current state of resources converted to SDA from four sources: ICSI-MR, IDIAP 1st series, U.Fribourg, ISSCO.
  • Joint IM2 Technical Report.
    • The IM2 Multimodal Meeting Browser Family
    • Lalanne D., Lisowska A., Bruno E., Flynn M., Georgescul M., Guillemot M., Janvier B., Marchand-Maillet S., Melichar M., Moenne-Loccoz N., Popescu-Belis A., Rajman M., Rigamonti M., von Rotz D. & Wellner P.
    • Abstract: The aim of this report is to describe the browsers that have been developed by various groups within the IM2 project, highlighting goals, design methodologies, key functionalities and evaluation methods used by each. The paper concludes with a tabular overview of the media, input and output modalities and special functionalities handled by each browser, as well as providing specific contact points and references.
    • PDF
  • Report IM2.MDM-15, November 2005.
    • Approaches to Topic Annotation of Discourse in Meeting Contexts
    • Eliane Lüthi
    • Abstract: This report aims to describe three approaches to topic annotation of discourse in meeting contexts: a derivational approach, a structure-based approach, and a keyword-oriented approach. It presents the advantages and disadvantages of each approach, as well as their usability for future automated annotation. It also discusses the results of a questionnaire designed to determine the need for topic annotation as it was first envisioned, and in view of these results it gives ideas for further work to be done in the domain.
      (20 pages)
    • PDF

 


Valid HTML 4.01!

Andrei Popescu-Belis
Last modified: June 2005