Past and future uses of text mining in ecology and evolution

Abstract

Ecology and evolutionary biology, like other scientific fields, are experiencing an exponential growth of academic manuscripts. As domain knowledge accumulates, scientists will need new computational approaches for identifying relevant literature to read and include in formal literature reviews and meta-analyses. Importantly, these approaches can also facilitate automated, large-scale data synthesis tasks and build structured databases from the information in the texts of primary journal articles, books, grey literature, and websites. The increasing availability of digital text, computational resources, and machine-learning based language models have led to a revolution in text analysis and natural language processing (NLP) in recent years. NLP has been widely adopted across the biomedical sciences but is rarely used in ecology and evolutionary biology. Applying computational tools from text mining and NLP will increase the efficiency of data synthesis, improve the reproducibility of literature reviews, formalize analyses of research biases and knowledge gaps, and promote data-driven discovery of patterns across ecology and evolutionary biology. Here we present recent use cases from ecology and evolution, and discuss future applications, limitations and ethical issues.

The Soil Food Web Ontology: aligning trophic groups, processes, resources, and dietary traits to support food-web research

“…The above applications of SFWO rely on our ability to compile large databases on the trophic ecology of soil consumers. Although much data is available from previously published research, the creation of literature-based datasets requires significant manual investment for literature searching, acquisition, screening, data extraction, and harmonisation of entities, such as species or trait names [29]. An alternative is to use information extraction approaches to automatically turn unstructured text into structured data, an approach currently taken by the Specialised Information Service Biodiversity Research 11 (BIOfid), which aims at extracting structured data from legacy Central European literature through semantic role labelling.…”

“…A number of the above applications of SFWO depend on our ability to collect large species-level databases on the trophic ecology of soil consumers. Although much data is available from previously published research, the creation of literature-based datasets requires significant manual investment for literature searching, acquisition, 6 https://github.com/biocodellc/ontology-data-pipeline 7 https://github.com/nleguillarme/inteGraph screening, data extraction, and harmonisation of entities, such as species or trait names (Farrell et al, 2022) A good ontology is consensual in nature, which means it should capture domain knowledge in a way that is accepted by the community. This is also key to a widespread adoption of a semantic model to the point it becomes a standard.…”

Patterns of host–parasite coinvasion promote enemy release and specialist parasite spillover

“…With current data availability, focusing on such details would strongly limit the 'across species' and 'across realms' aims of our study, compromising the potential to provide generalized results. However, we believe these approaches could be valid in the near future as techniques to mine host-parasite data improve (Farrell et al, 2022).…”

Effect of IUCN Red List category on public attention to mammals

“…Culturomics (Ladle et al., 2016; Correia et al., 2021a, b) and text mining (Farrell et al., 2022) have become increasingly useful tools for analyzing conservation issues. However, these data can often be complex, and there are several limitations to my data.…”