Ifla art of life presentation final

IFLA Section on Science & Technology August 19,2014

Merging the Worlds of Art and Science

Trish Rose-SandlerCenter for Biodiversity Informatics, Missouri Botanical Garden, St. Louis, MO, USA

Nancy E. GwinnSmithsonian Libraries, Washington, DC, USA

Constance RinaldoErnst Mayr Library, Museum of Comparative Zoology, Harvard University,Cambridge, MA, USA

What is Biodiversity Content?

Digital Repository ofBooks

Field Notebooks

Journals

• Extensive

• Open

• Global

• Linked

Digital Library

78,947 Titles

141,770 Volumes

44,275,666 Pages

What is the Biodiversity Heritage Library?

• American Museum of Natural History Library• California Academy of Sciences Library• Cornell University Library• Harvard University Botany Library• Ernst Mayr Library of the Museum of Comparative

Zoology• Library of Congress• Marine Biological Laboratory Library• Missouri Botanical Garden Library• Natural History Museum, London Library• The New York Botanical Garden Library• Royal Botanic Garden, Kew, Library• Smithsonian Libraries• Washington University at St. Louis Library• University of Illinois Library• United States Geological Survey Libraries

15MEMBERS

4AFFILIATES

BHL Central

• Academy of Natural Sciences Library• Chicago Botanical Garden Library• The Field Museum Library• Los Angeles County Natural History Museum Library

Global Partners

Europe

China

Australia

Egypt

Brazil

Africa

Singapore

• Scientific descriptions of animals, plants, nature in general (Taxonomic literature)

• Type in a taxon name and find all detected occurrences in the text corpus.

• Systematic biology is easier for scientists—the taxonomic impediment is resolved.

BHL = Libraries + Technology + Science

BHL Book Viewer

ChallengesMillions of natural history illustrations

Very few have metadata that describe their content or where they are in the book

BHL and Flickr—90,000 Images

Only pretty pictures

Still too manual

Many illustrations not included

o Full title—The Art of Life: Data Mining and Crowdsourcing the Identification and Description of Natural History Illustrations from the Biodiversity Heritage Library (BHL)

o Grant given to Missouri Botanical Garden in St. Louis, Missouri

o Funded by U.S. National Endowment for the Humanities

o Runs May 2012-April 2015

Art of Life?

o Define metadata schema for natural history illustrations

o Build algorithms to automate identification of illustrations

o Enhance existing tools to classify illustrations

o Crowdsource adding descriptive metadata

o Integrate metadata into BHL and share images and metadata more widely

Goals/Objectives

Algorithms to automate identification of images tested:

Picture blocks 87-88% effectiveContrast 87-88% effectiveColor ineffectiveCompression ineffective

How to Identify Illustrations?

Web Application

For analyzing & visualizing algorithm results

Basic Workflow

Macaw

An interface for classifying pages identified by algorithms

Crowdsourcing

Flickr tagging

Wikimedia

Commonsapply schema

For adding description of illustrations

Describing Natural History Illustrations

Schema = VRA Core with elements from Darwin Core

o Running algorithms across BHL corpus

o 1.5 million pages processed

o 300,000 pages with images

o Estimate will be 15% of corpus—pages with images

o Classifying results—78,000 pages so far

o Testing bulk upload to Wikimedia Commons

o Testing extraction of metadata from tool

o BHL architecture modified and ready to store and preserve newly created metadata

Art of Life Status

o Images serve multiple audienceso Scientistso Artistso Historianso Teachers

o Rich cultural heritage contento If specimens are gone, may be the only way to

document a specieso Image finding algorithms and analysis tools on

githubgithub.com/IMAmuseum/artoflife

o Image classification tool (Macaw) on githubgithub.com/cajunjoel/macaw-neh

Benefits

For more information: Trish Rose-Sandler, Principal [email protected]/Art+of+Life

Questions?