This research aims to help scientists develop and use relatively simple tools to describe species in a way that makes those descriptions easier to share with other scientists and easier for computers to process and analyze. Taxonomists are scientists who describe the world's biodiversity. Taxonomists' descriptions of millions of species allow scientists to do many different kinds of research, including basic biology, environmental science, climate research, agriculture, and medicine. The problem is that describing any one species is not easy. The language used by taxonomists to describe their data is complex, and typically not easily understandable by computers nor even other scientists. This situation makes it difficult to search for patterns across the millions of species that have been documented by thousands of different researchers over many decades of work worldwide. Innovation from this project is applicable to the long-term development of open source software initiatives serving laboratories throughout the world, and the research facilitates the production of open, shared data, as mandated by various federal agencies. As a result of this project, these data will become more accessible and informative to the general public. The project provides rich, real-world training for graduate students in library and information sciences, training them to be cross-disciplinary researchers in a field that is in need of new experts. Collaborating experts studying bees, wasps, and ants will receive training on the cutting edge theories and methods from the bioinformatics toolbox developed as a consequence of this project. In return their contributions of data will act as the basis for computational benchmarks needed in areas of logical inference and data modeling.
This research addresses the problem of how to produce and utilize semantic data, specifically semantic phenotypes, within the taxonomic context of describing the Earth's biological diversity. The approach to be taken is bottom-up and iterative, involving the rapid prototyping of tools, combining of existing tools, and the tailoring of applications developed for one purpose but now being reused for this scientific activity. Scientists are busy innovating partial solutions by tinkering with and combining available computer programs and datasets. Their efforts comprise an incredibly productive source of innovation, since it is often much easier and faster to combine computer resources that already exist than to build something from scratch. However such cobbling together of resources to meet a need can benefit from analysis and active support. In particular, a more principled set of approaches can make innovations easier to share and to maintain. With a focus on the Hymenoptera, the researchers plan an innovative approach for biodiversity informatics based on work in the field of Computer Supported Cooperative Work (CSCW). Using a combination of ethnography to define work practice, user-centered design, and iterative agile software development, the collaboration between information scientists, biologists, and application developers aims to produce a suite of concrete deliverables, a rapid prototype portfolio, comprising interface and workflow tools, and end user requirements for semantic phenotype production. The project will explore and document examples of innovative prototyping of solutions by scientists to understand how it occurs, what it is that scientists most need, and how these can be most effectively supported. These components may be generalized to allow broader scientific use.
|Effective start/end date||6/15/14 → 5/31/16|
- National Science Foundation: $192,790.00