Training session in the framework of the EMODnet Biology Phase IV on the feasibility study for recognition of specific ecological traits and sampling devices/methodologies in text

EMODnet Biology is the EU service for European in-situ marine biodiversity data. It was established in 2009, following the publication of the Maritime Policy Blue Book in 2007. Its main objectives include the curation, quality control and publication of marine biodiversity data in European marine regions and the Caribbean seas where European countries have overseas territories or by European scientists and/or institutions outside of the European area. EMODnet Biology focuses not only on recent but also historical data, i.e. data that have been collected over 50 years ago. Part of the work is also dedicated to creating interoperable and reproducible products based on open data and engaging with the regional sea conventions, EU and international reporting mechanisms.

The training session is organised in the framework of EMODnet Biology and in particular on the ongoing Feasibility Study on Ecological Traits and sampling devices/methodologies identification with text mining. This training session focuses on the interdisciplinary field of text mining and its application on marine ecological traits. During the session, participants will receive an overview of state-of-the-art text mining technologies and bioinformatic tools. A balanced approach of lectures and practical sessions will provide a general background and the latest advancements in the field as well as its relationship with other disciplines.

The session will start with training the participants about literature curation workflows, standards and resources. Participants will prepare their own literature and texts and follow the workflow to extract information on traits and species names. Using machine learning algorithms and tools they will evaluate the results and provide feedback on user interface, usability and feature requests. Thus, the participants have a unique opportunity to work on the complete workflow from literature preparation to data analysis. This session will also address the annotation of texts to improve the current models. Programming is a key part of the session, during which an introduction on how to process data from raw text to the application of algorithms for text mining, will take place. Trainers will provide short lectures with follow-up hands-on exercises in order to have the participants learning and assessing their knowledge. Registration via: info@advancesvs.com