TI Methods Speaker Series: Keep a human in the machine and other lessons learned from deploying and maintaining Colandr

photo of Samantha Cheng

TI Methods Speaker Series: Keep a human in the machine and other lessons learned from deploying and maintaining Colandr

photo of Samantha ChengThis live webinar has already taken place. Scroll down to view the video recording.

While methods to systematically identify and synthesize evidence from the literature in an transparent and systematic fashion have tremendous potential to facilitate evidence-based decision making, these require manual searches that are both labor-intensive and prone to human error. Technical advances in human-assisted machine learning and data visualization have the potential to substantially reduce the effort required to find the proverbial “needles” in a haystack of tens of thousands of documents. Colandr, an open-source web-based platform for conducting evidence syntheses and reviews, employs two machine learning and natural language processing processes to improve the efficiency of finding relevant citations and improving data extraction. Colandr was initially created as a DataKindDataCorps project in collaboration with a Science for People and Nature Partnership (SNAPP) Working Group on Evidence-Based Conservation and Conservation International to develop an open-source method for updating large-scale systematic evidence maps. Since its launch, Colandr hosts over 2,000 projects across a wide-range of disciplines from environmental management to health to education. In this talk, we discuss how Colandr can improve efficiency for syntheses in multidisciplinary topics, using nature-based conservation as a case study. We also discuss the opportunities and challenges of developing an open source machine learning platform for the evidence synthesis community. Open source software – i.e. software with source code that is publicly accessible for inspection, use, and modification – have been touted as a reproducible and sustainable answer to these resource constraints. Open source approaches are centered on sharing and collaboration as a means to produce stable, supported, and advanced software for greater use. While open source presents several opportunities for creating sustainable and reproducible evidence syntheses and derived products, there are many challenges that should be considered.


WHEN: Wednesday, June 29th, 2022 from 12:00 to 1:00 PM PDT [convert to your local time]

WHERE: Offered online using the Zoom platform.

TITLE: Keep a human in the machine and other lessons learned from deploying and maintaining Colandr

SPEAKER: Samantha Cheng, Ph.D., Director of Conservation Evidence at the World Wildlife Fund.


About the speaker: Samantha Cheng is the Director of Conservation Evidence at the World Wildlife Fund and works across the organization and with external organizations to advance evidence-informed practice in conservation programs. She is an interdisciplinary conservation scientist whose work draws on biological, social, and computer sciences, to understand connections between nature and human well-being and build tools and assessments for evidence synthesis and monitoring and evaluation and learning. She runs two online open access tools to facilitate assessing and accessing evidence—Colandr and the Evidence for Nature and People Data Portal.


About the TI Methods Speaker Series: The TI Methods Speaker Series are offered free of charge and everyone is welcome. The event is held at noon on the last Wednesday of each month. During the COVID-19 pandemic, while physical distancing measures are in effect, the TI Methods Speaker Series are offered via videoconference. The presentations are recorded and the video recordings are posted online. Click here to view the scheduled topics for 2022 and click here to view a list of TI Methods Speaker Series talks offered in 2021.

No Comments

Post A Comment