Workshop  

Text Mining for the BioCuration Workflow


Organizers:
Lynette Hirschman, MITRE: lynette@mitre.org
Gully APC Burns, ISI/USC: GullyBurns@gmail.com
K. Bretonnel Cohen, University of Colorado: kevin.cohen@gmail.com
Martin Krallinger, CNIO: mkrallinger@cnio.es
Cathy Wu, Georgetown: wuc@georgetown.edu


The goals of this workshop are to update the BioCurator community on the state of the art in text mining and to elicit the requirements from the BioCurator community for enhanced tools to support the curation workflow.

The workshop will be divided into two parts. The first part will be tutorial in nature and will cover what tools are available, how to integrate components into a curation workflow, and what kind of performance to expect based on available resources. We will also discuss models for curation, including structured digital abstracts.
The second part of the workshop will be interactive, with a focus on understanding the diversity of curation workflows and requirements. For this part, we will invite participants to submit short presentations (5-10 min) on their requirements and their experiences or needs integrating text mining into their curation workflow. We will also discuss how to create partnerships between the bio-text mining tool developers and the BioCurator community.


It is hoped that the workshop will have as its outcome a statement of requirements for text mining tools and capabilities needed to support the BioCuration workflow. This set of requirements can be taken forward to the bio-text mining community to encourage partnerships and further discussion, for example at the ISMB BioLINK SIG meetings, and to focus challenge evaluation activities such as BioCreative.


Tutorial
  • Understanding text mining technology
    • Document retrieval for triage and curation prioritization
    • Information extraction: entity tagging & normalization, relation extraction
    • Visualization tools
    • Requirements for interactive interfaces
    • Integration into the workflow
  • Models of curation
    • Post-publication curation (the current model)
    • Direct author submission (cost-benefit tradeoffs)
    • Publisher aids (structured digital abstracts and relation to curation)
Understanding the Curation Workflow
  • Presentation of use cases and success stories
  • Capture of BioCuration requirements
  • Discussion of how to create partnerships

Participants interested in making a short presentation should send their abstract directly to Lynette Hirschman by January 15, 2009.

Please use the form for workshop/tutorial registration to register for this workshop. The registration is free of charge. Workshop participants have to be registered for the Biocuration Conference.


Workshop Agenda
[PDF]

Friday, April 17, 2009



SESSION I

14:00-14:15

Lynette Hirschman
Introduction to the Workshop
14:15-14:30 K. Bretonnel Cohen
Text Mining Tutorial
14:30-15:00 Gully APC Burns, Martin Krallinger
A Framework for BioCuration Workflows
15:00-15:15 Discussion
15:15-15:30 Talk: Lourenço, Carneiro, Carreira, Rocha, Rocha, Ferreira
"Bringing Text Miners and Biologists Closer Together"
15:30-15:45 Talk: Bada, Eckert, Garcia, Evans, Sitnikov, Baumgartner, Ogren, et al
"The Colorado Richly Annotated Full-Text (CRAFT) Corpus: A Resource for Biocurational Text-Mining Research"
15:45-16:15 Break


SESSION II

16:15-16:30

Talk: Veuthey, Pillet, Yip, Ruch
"Text Mining for Swiss-Prot Curation: A Story of Success and Failure"
16:30-16:45 Talk: Wiegers, Davis, Hirschman, Cohen, Rosenstein, Mattingly
"Developing a Text Mining Prototype for the Comparative Toxicogenomics Database Biocuration Project"
16:45-17:00 Talk: Dowell, McAndrews-Hill, Hill, Blake
"Evaluating Text Mining Tools for the Biocuration Workflow at MGI"
17:00-17:15 Talk: Chatr-aryamontri, Licata, Ceol, Cesareni
"Structured Digital Abstracts: the FEBS Letters Experiment"
17:15-18:00 Lynette Hirschman
Discussion
- Feedback on BioCuration Workflow
- Building a Roadmap for Text Mining Support for BioCuration

    Back to Conference Home Page