Abstract

The electronic availability of biomedical publications has led to a surge of interest in using text mining (TM) as a way of accessing this ever expanding knowledge store.

It is clear that future advances in understanding of biomedical issues will rely on the discovery of interconnections of data and models, which are as yet unrecognised. TM will provide a major contribution to joining up diverse experimental evidence and will give new insights into biological processes.

Assembling a picture of the mechanisms of life in a systems or pathway context, for example, can be greatly facilitated by the use of tools such as intelligent search and knowledge distillation over very large document collections.

However, effective tools require cooperation between the designers and the users. This workshop aims to bring the biomedical and TM communities together by showcasing the possible applications of TM while opening up dialogue that will facilitate specification of use cases to drive the next generation of applications.

Key Questions

Some key questions that the workshop will address are:

• Is there such a thing as an ultimate TM application or are individual solutions needed for each problem?
• What are the common difficulties in eliciting user requirements and what are the solutions?
• Is TM really the way forward, or are higher-precision hand-curated databases and ontologies more practical?
• What are the specific challenges for TM in the biomedical domain?
• How can TM technology be successfully incorporated into research and curation workflows despite being imperfect?

Workshop Flyer

Also see the workshop poster.