Fast, scalable, and automated identification of articles for biodiversity and macroecological datasets

Cornford, Richard; Deinet, Stefanie; Palma, Adriana De; Hill, Samantha L. L.; McRae, Louise; Pettit, Benjamin; Marconi, Valentina; Purvis, Andy; Freeman, Robin

doi:10.1111/geb.13219

Cited by 21 publications

(26 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In NLP, the sub-field of information retrieval develops search algorithms and models that suggest articles of potential interest. In a recent ecological application, Cornford et al (2021) [38] train machine learning models to classify literature as relevant to the PREDICTS database [39], a literature-based database of biodiversity responses to human impacts. Their best models could distinguish relevant from non-relevant articles with over 90% accuracy based only on title and abstract text, significantly improving the speed and ease with which new articles can be screened for database inclusion.…”

Section: Expanding Literature-based Datasetsmentioning

confidence: 99%

Past and future uses of text mining in ecology & evolution

Farrell¹,

Brierley²,

Willoughby³

et al. 2022

Preprint

View full text Add to dashboard Cite

Ecology and evolutionary biology, like other scientific fields, are experiencing an exponentialgrowth of academic manuscripts. As domain knowledge accumulates, scientists will neednew computational approaches for identifying relevant literature to read and include informal literature reviews and meta-analyses. Importantly, these approaches can alsofacilitate automated, large-scale data synthesis tasks and build structured databases fromthe information in the texts of primary journal articles, textbooks, grey literature, andwebsites. The increasing availability of digital text, computational resources, andmachine-learning based language models have led to a revolution in text analysis andNatural Language Processing (NLP) in recent years. NLP has been widely adopted acrossthe biomedical sciences, but is rarely used in ecology and evolutionary biology. Applyingcomputational tools from text mining and NLP will increase the efficiency of data synthesis,improve the reproducibility of literature reviews, formalize analyses of research biases andknowledge gaps, and promote data-driven discovery of patterns across ecology andevolutionary biology. Here we present recent use cases from ecology and evolution, anddiscuss future applications, limitations, and ethical issues.

show abstract

Section: Expanding Literature-based Datasetsmentioning

confidence: 99%

Past and future uses of text mining in ecology & evolution

Farrell¹,

Brierley²,

Willoughby³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…These open-access databases will provide GRIN with millions of data points spanning the globe. An increasing amount of data can also be identified (Cornford et al 2021) and downloaded from the published literature (e.g., Grilo et al 2018, Miranda et al 2019, Pereira et al 2019). GRIN will continue to incorporate outside data and mobilize legacy datasets.…”

Section: The Precursors To Grinmentioning

confidence: 99%

Commentary: the Past, Present, and Future of the Global Raptor Impact Network

McClure

Anderson

Buij

et al. 2021

Journal of Raptor Research

View full text Add to dashboard Cite

Most raptor populations are declining and nearly a fifth are threatened with extinction; thus there is a need to increase collaboration to ensure efficient and effective research, management, and conservation. Here, we introduce the Global Raptor Impact Network (GRIN; www.globalraptors.org), a tool to enhance collaboration and conservation impact of the raptor research community. We provide an overview of the history and current state of GRIN, including plans for expansion. Predecessors to GRIN include The African Raptor DataBank, which was launched in 2012 to ascertain the conservation status of raptors across Africa; and the Global Raptor Information Network, which was launched in the late 1990s as a website to provide information regarding diurnal raptors and facilitate communication among researchers. GRIN expands the data collection and storage capabilities of the African Raptor DataBank to a global scale via mobile application. We have implemented data-sharing rules to ensure the safety of sensitive species, and users of the GRIN mobile app can designate their records as confidential. GRIN staff and partners are developing analyses of species' population trends and geographic distributions to aid in conservation assessments. GRIN is also developing systematic reviews, detailed bibliographies, and online accounts that will summarize the state of knowledge for each raptor species. We hope that GRIN will benefit the entire raptor research community and aid in the collaboration necessary to help raptor populations thrive in the Anthropocene.

show abstract

“…'Big data' approaches, and associated computational tools, provide a means to wrangle the extensive ecological literature into usable information (Westgate et al 2018). Much of the recent development in synthesis methods has been in expediting and automating the searching for (Grames et al 2019), and screening of (Wallace et al 2012, Shackelford et al 2020, Cornford et al 2021, papers to address research questions. Within the medical literature, some approaches have even managed to automate the entire systematic review procedure (Marshall and Wallace 2019, Gates et al 2020, Marshall et al 2020, Yang et al 2020, Brassey et al 2021).…”

Section: Introductionmentioning

confidence: 99%

Automated synthesis of biodiversity knowledge requires better tools and standardised research output

et al. 2022

Self Cite

View full text Add to dashboard Cite

As the impact of anthropogenic activity on the environment has grown, research into biodiversity change and associated threats has also accelerated. Synthesising this vast literature is important for understanding the drivers of biodiversity change and identifying those actions that will mitigate further ecological losses. However, keeping pace with an ever‐increasing publication rate presents a substantial challenge to efficient syntheses, an issue which could be partly addressed by increasing levels of automation in the synthesis pipeline.Here, we evaluate the potential for automated tools to extract ecologically important information from the abstracts of articles compiled in the Living Planet Database. Specifically, we focused on extracting key information on taxonomy (studied species names), geographic location and estimated population trend, assessing the accuracy of automated versus manual information extraction, the potential for automated tools to introduce biases into syntheses, and evaluating if synthesising abstracts was enough to capture the key information from the full article.Taxonomic and geographic extraction tools performed reasonably well, although information on studied species was sometimes limited in the abstract (compared to the main text) preventing fast extraction. In contrast, extraction of trends was less successful, highlighting the challenges involved in automating information extraction from abstracts, such as deficiencies in the algorithms, linguistic complexity associated with ecological findings, and limited information when compared to the main text.In light of these results, we cautiously advocate for a wider use of automated taxonomic and geographic parsing tools for ecological synthesis. Additionally, to further the use of automated synthesis within ecology, we recommend a dual approach: development of improved computational tools to reduce biases; and enhanced protocols for abstracts (and associated metadata) to ensure key information is included in a format that facilitates machine‐readability.

show abstract

Fast, scalable, and automated identification of articles for biodiversity and macroecological datasets

Abstract: This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Cited by 21 publications

References 40 publications

Past and future uses of text mining in ecology & evolution

Past and future uses of text mining in ecology & evolution

Commentary: the Past, Present, and Future of the Global Raptor Impact Network

Automated synthesis of biodiversity knowledge requires better tools and standardised research output

Contact Info

Product

Resources

About

Fast, scalable, and automated identification of articles for biodiversity and macroecological datasets

Abstract: This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Cited by 21 publications

References 40 publications

Past and future uses of text mining in ecology &amp; evolution

Past and future uses of text mining in ecology &amp; evolution

Commentary: the Past, Present, and Future of the Global Raptor Impact Network

Automated synthesis of biodiversity knowledge requires better tools and standardised research output

Contact Info

Product

Resources

About

Past and future uses of text mining in ecology & evolution

Past and future uses of text mining in ecology & evolution