Abstract
Scientific articles often contain relevant geographic information such as where field work was performed or where patients were treated. Most often, this information appears in the full-text article contents as a description in natural language including place names, with no accompanying machine-readable geographic metadata. Automatically extracting this geographic information could help conduct meta-analyses, find geographical research gaps, and retrieve articles using spatial search criteria. Research on this problem is still in its infancy, with many works manually processing corpora for locations and few cross-domain studies. In this paper, we develop a fully automatic pipeline to extract and represent relevant locations from scientific articles, applying it to two varied corpora. We obtain good performance, with full pipeline precision of 0.84 for an environmental corpus, and 0.78 for a biomedical corpus. Our results can be visualized as simple global maps, allowing human annotators to both explore corpus patterns in space and triage results for downstream analysis. Future work should not only focus on improving individual pipeline components, but also be informed by user needs derived from the potential spatial analysis and exploration of such corpora.
Highlights
Geographical information permeates the written world, appearing as place names or place descriptions in texts including news articles, blog posts, social media content, historical documents, and scientific articles
Research on extracting geographical information from text has often focused on news articles [1,2,3] and social media content [4,5,6], with surprisingly limited attention being directed towards the increasing number of published scientific articles
We develop a fully automatic pipeline which starts from a collection of scientific articles and their pipeline recall and F1. (PDF) and outputs a set of location strings and their sentence context, as well as structured information and a geometric representation for each string (Fig 1)
Summary
Geographical information permeates the written world, appearing as place names or place descriptions in texts including news articles, blog posts, social media content, historical documents, and scientific articles. With each passing year, scientists face an ever-growing stack of scientific articles to sort through, read, understand, and build upon Many of these articles contain important geographical information: perhaps soil samples were taken from a certain region, patients were treated in a particular hospital, or interviews were conducted in a village or neighborhood. Researchers must manually sift through article contents to identify any relevant locations, a time-consuming process. Linking these textual place descriptions to spatial representations (such as point coordinates, a bounding box, or a polygonal region) requires significant additional work and should ideally respect the scale and precision of locations described in the text. Good performance is obtained after adding custom pre- and post-processing steps, such as enhancing word lists with geology-specific terms and detecting citations in order to skip them as location candidates
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
