Customisable Curation Workflows in Argo
暂无分享,去创建一个
We adapt Argo, a Web-based text mining workbench, to accommodate a curation task for BioCreative IV Track 5 User Interactive Task. The system allows users to build and run custom processing workflows composed of a subset of available elementary processing components. Running such workflows results in the generation of annotations for each input document. Argo has the capability to include a manual annotation editor as part of any workflow and therefore facilitates manual inspection and correction of otherwise automatically produced annotations. As a case study, we propose a curation task that involves the annotation and identification (normalisation) of concepts relevant to metabolic processes, including chemicals, gene or gene products, and trigger/bioprocess expressions.
[1] Sophia Ananiadou,et al. Using Workflows to Explore and Optimise Named Entity Recognition for Chemistry , 2011, PloS one.
[2] Dietrich Rebholz-Schuhmann,et al. Biological network extraction from scientific literature: state of the art and challenges , 2014, Briefings Bioinform..
[3] Nigel Collier,et al. Introduction to the Bio-entity Recognition Task at JNLPBA , 2004, NLPBA/BioNLP.
[4] Sophia Ananiadou,et al. Making UIMA Truly Interoperable with SPARQL , 2013, LAW@ACL.