This visual search interface was developed within the context of the Mining Biodiversity project (http://miningbiodiversity.org/).

This project aims to transform the Biodiversity Heritage Library (BHL) into a next-generation social digital library resource to facilitate the study and discussion of legacy science documents on biodiversity by a worldwide community. The project integrates novel text mining methods, visualisation, crowdsourcing and social media into the BHL.

How to cite this work:

Nhung T.H. Nguyen, Georgios Kontonatsios, Riza Batista-Navarro, Axel J. Soto, Sophia Ananiadou. "Constructing a Biodiversity Terminological Inventory". Under review.

Summary

This web application searches for documents on the Biodiversity Heritage Library (BHL). Unlike typical search systems, this application suggests synonymous or semantically related terms that can be used to expand the query. The suggested terms are obtained by: (1) available taxonomies, e.g., Catalogue of Life and Encyclopedia of Life; (2) extracting semantically related terms from BHL by using distributional semantic methods.

Motivation

The motivation behind such automatic query expansion comes from showing the user not only documents matching the original query but also those which mention names that are semantically relevant, e.g., other species sharing similar habitat or taxonomic classification. For instance, when a user searches for documents pertaining to “rock pigeon”, it might be useful for him/her to find documents mentioning “columba livia”—its scientific name—or “speckled pigeon”—another pigeon belonging to the same genus. This feature was implemented as part of our aim to help a BHL user broaden the coverage of his/her search while hopefully keeping unwanted, noisy results to a minimum.