Mining Biodiversity: Enriching Biodiversity Heritage with Text Mining and Social Media

Mining Biodiversity is one of the projects that won in the third round of the transatlantic Digging Into Data Challenge, a competition aiming to promote the development of innovative computational techniques that can be applied to big data in the humanities and social sciences. Overall, 14 teams representing collaborations from Canada, US, UK and the Netherlands have been selected. Four of the selected projects come from the UK.

The project is an international collaboration between the National Centre for Text Mining (UK), Missouri Botanical Garden (US), Dalhousie University's Big Data Analytics Institute (Canada) and Ryerson University's Social Media Lab (Canada). NaCTeM was also a recipient of the 2011 Digging into Data call with the Integrated Social History Environment for Research (ISHER) project.

Summary

The Mining Biodiversity project aims to transform the Biodiversity Heritage Library (BHL) into a next-generation social digital library resource to facilitate the study and discussion (via social media integration) of legacy science documents on biodiversity by a worldwide community and to raise awareness of the changes in biodiversity over time in the general public. The project integrates novel text mining (TM) methods, visualisation, crowdsourcing and social media into the BHL. The resulting digital resource will provide fully interlinked and indexed access to the full content of BHL library documents, via semantically enhanced and interactive browsing and searching capabilities, allowing users to locate precisely the information of interest to them in an easy and efficient manner.

Project goals

By promoting the development of capabilities that will foster collaboration amongst researchers from the fields of History of Science, Environmental History, Environmental Studies, Library and Information Science and Social Media, the proposed project will make a significant impact on the above disciplines by:

enriching a large-scale library, i.e., the BHL, via innovative application of text mining techniques to produce semantic metadata and a term inventory,
providing improved access to biodiversity-related digital artifacts via an enhanced search engine and visualisation of results, and
stimulating increased collaboration, interaction and sharing of information amongst BHL users via the social media environment.

Details

The project has five major components:

Automatic correction of errors in text extracted automatically from legacy biodiversity literature via optical character recognition (OCR).
Development of a crowdsourcing facility that will encourage users to annotate legacy texts with semantic metadata.
Adaptation of text mining technologies to extract metadata (i.e., terminology, entities and significant events) automatically and to track terminology evolution over time. This will facilitate semantic search, allowing users to explore search results according to multiple information dimensions or facets.
Interactive visualisation techniques will be used to help users to make sense of search results through the integration of next generation browsing capabilities, assisted by a semantic similarity network of important terms and entities.
Design of a social media layer, serving as an environment for diverse users to interact and collaborate on science, public education, awareness and outreach.

Architecture

Project information

The project ran from March 2014 until September 2015, and was funded by AHRC, ESRC, Innovation.ca, Institute of Museum and Library Services, JISC and NEH.

Publications

Batista-Navarro, R., Hammock, J., Ulate, W. and Ananiadou, S. (2016). A Text Mining Framework for Accelerating the Semantic Curation of Literature. In: Proceedings of the 20th International Conference on Theory and Practice of Digital Libraries (TPDL 2016), pp. 459-462, Springer

Batista-Navarro, R., Soto, A., Ulate, W. and Ananiadou, S. (2016). Text Mining Workflows for Indexing Archives with Automatically Extracted Semantic Metadata. In: Proceedings of the 20th International Conference on Theory and Practice of Digital Libraries (TPDL 2016), pp. 471-473, Springer

Talks

A talk entitled Enriching the legacy literature with OCR corrections and text-mined semantic metadata at the Annual Conference of Biodiversity Information Standards (TDWG) 2014 in Jönköping, Sweden.

A talk entitled Unlocking knowledge in biodiversity legacy literature through automatic semantic metadata extraction at the Annual Conference of Biodiversity Information Standards (TDWG) 2015 in Nairobi, Kenya.

A tutorial entitled Text mining workflows for indexing archives with automatically extracted semantic metadata at the 20th International Conference on Theory and Practice of Digital Libraries in Hannover, Germany.

A talk entitled Expanding Access to Biodiversity Literature at the 48th Annual Meeting of the Council on Botanical and Horticultural Libraries in Cleveland, Ohio