Samantha Cheng
I am a biodiversity scientist at the Center for Biodiversity & Conservation at the American Museum of Natural History. My interests are broadly in determining how to improve the process of using scientific evidence and evidence synthesis in conservation decision-making from an interdisciplinary approach. I specifically focus on how data science and new technologies (incl. computer science and genomics) can improve the process.
Current initiatives include:
1) Evidence for Nature and People Data Portal - this is an interactive, open access data portal and visualization tool which allows users to find, filter and explore data from a systematic map on the impacts of conservation interventions on human well-being (McKinnon et al. 2016). Users can visualize the data spatially, on a summary dashboard, and download subsets of the full dataset and bibliography. This portal is intended to host multiple datasets from evidence synthesis studies that examine links between nature and people.
2). Colandr - this tool is developed in collaboration with a team of volunteer data scientists from DataKind, a non-profit data science organization. Colandr utilizes machine learning and natural language processing algorithms to find relevant studies based on desired parameters and keywords as well as aids in automated data extraction. This is currently in beta testing.
Current initiatives include:
1) Evidence for Nature and People Data Portal - this is an interactive, open access data portal and visualization tool which allows users to find, filter and explore data from a systematic map on the impacts of conservation interventions on human well-being (McKinnon et al. 2016). Users can visualize the data spatially, on a summary dashboard, and download subsets of the full dataset and bibliography. This portal is intended to host multiple datasets from evidence synthesis studies that examine links between nature and people.
2). Colandr - this tool is developed in collaboration with a team of volunteer data scientists from DataKind, a non-profit data science organization. Colandr utilizes machine learning and natural language processing algorithms to find relevant studies based on desired parameters and keywords as well as aids in automated data extraction. This is currently in beta testing.
Publications and Links
See website