CLEAR is the Computational Language and EducAtion Research institute at the University of Colorado Boulder. The CLEAR institute is a leader in computational linguistics, with funding from the biomedical field, NSF and DARPA.

The ClearEarth project is a National Science Foundation-sponsored initiative that brings together computational linguists and scientists who work on the earth surface system - scientists from the INSTAAR-CSDMS, the Ronin Institute for Independent Scholarship, and the Biodiversity Data Integration Group at the Research Data Alliance. The goal of the project is to use annotation, natural language processing (NLP), and machine learning (ML) tools to bring about advances in three distinct domains: Geology, Cryology and Ecology. In each of the domains - earth, ice, life - there is a goal of establishing semantic (language/text) information resources which can help with issues like ‘big data’, ‘dark data’, data discovery and mining, semantic alignment, and query mediation. The reason for this is the need to characterize Earth as a single integrated system of geology, climate and biology. The hoped-for effect is access to the information that is held in texts, and thereby an acceleration of research results over a widened ‘discovery space’. The NSF award for this project can be viewed at Porting Biomedical Semantic Technologies to the Earth Sciences.

The goal of the hackathon is to tackle cross-domain - earth sciences / computational linguistics - problems with the help of local experts, linguists, and earth scientists. Those problems could involve natural text, datasets, software tools, visulizations. Applicants must propose a week-long project as part of their application. Suggestions are: NLP/ML tools, information extraction, ontology-building, or any topic of interest that relates linguistics and earth science in a ML framework. In the earth sciences we include all earth-ice-life (geo-cryo-eco) studies. At the end of the hackathon, participants will present their tools and post them on GitHub. Applicants may apply as individuals or in groups of up to three - building a team within or across universities is strongly encouraged! Applicants may also choose to form a small group at the outset of the hackathon, or may choose to work alone. The only requirement is the development of a prototype linguistics-related software or data tool. The hackathon is NSF funded, and applicants need only bring a laptop and other essentials. Accommodation, travel and transport, and meals+ are covered. Other benefits of attending the hackathon include:

Image credit: Flickr/HackNY.org

• Expanding your professional network
• Meeting industry and research experts in computational linguistics
• Developing a tool to accelerate your research
• Working in groups or small teams
• Gaining experts' feedback on your project
• Connecting with the CU-Boulder CLEAR team and with linguistics researchers from across the country
• Earning graduate-level course credit through CU-Boulder (TBD)
• Enjoying intense collaborative coding, unlimited pizza and Red Bull, a free stay in Boulder, and the Rocky Mountains nearby