Poster Open Access
Bohuon, Jean-Baptiste; Ba, Mouhamadou; Chaix, Estelle; Bossy, Robert; Nédellec, Claire
Alvis is a high-performance Text and Data Mining (TDM) suite, tailored to the needs of information extraction from scientific publications in Life Science and Agriculture. The Alvis suite is developed at Bibliome, an INRA team specialized in the development of information extraction, ontology building and TDM methods. The Alvis library of modules include state of the art entities and relations annotation modules, some of which use machine learning algorithms and semantic resources. Application design with Alvis is done by the combination and adaptation of many components to the task in the form of workflows. Alvis is one of the TDM platform solutions (with Gate and UIMA) that are being integrated by the OpenMinTeD community to create an open, service-oriented e-Infrastructure for TDM of scientific publications. OpenMinTeD is a European project that aims to develop a platform to foster and facilitate the use of TDM technologies by the scientific community. The OpenMinTeD community is dedicated to developing workflows enabling non-experts computer scientists to apply state of the art TDM workflows. Galaxy is used by OpenMinTeD as the workflow engine and is associated to a resource registry and user interfaces within the platform. This platform enables input data, such as corpora and ontologies to be tighly associated to specific workflows. The Bibliome team is currently integrating Alvis modules as Galaxy tools into the OpenMinTeD platform. This poster specially presents the Alvis approach, how it is adapted to OpenMinTeD and how it is being integrated to Galaxy through specific use cases.
Ba et al. (2016). Interoperability of corpus processing workflow engines: the case of AlvisNLP/ML in OpenMinTeD. Presented at Meeting of working Group Medicago sativa, Portoroz, SVN.
Chaix et al. (2016). Overview of the regulatory network of plant seed development (SeeDev) task at the BioNLP shared task 2016. In: Proceedings of the 4th BioNLP Shared Task Workshop (p. 12-22). Presented at BioNLP Shared Task, Berlin, DEU. Stroudsburg, USA : The Association for Computational Linguistics.