Extracting and Aggregating Hierarchical Toponyms in Abstracts of Scientific Articles in Urban Studies
Tianye Ren (TU Delft - Architecture and the Built Environment)
Nan Bai (TU Delft - Architecture and the Built Environment)
Ana Pereira Roders (TU Delft - Architecture and the Built Environment)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
This study introduces a literature review tool for abstract-screening in urban studies. It presents two pipelines that focus on linking geoparsing outputs and hierarchical toponym aggregation. Pipeline1 uses batched matching with GeoNames for fast but coarse aggregation. Pipeline2 enhances the accuracy by incorporating a toponym resolution model to address geo/geo ambiguity, and semantic checks to correct potential resolution errors. Evaluated on 500 abstracts, Pipeline2 achieves a precision of 0.96 and a recall of 0.98 for aggregated toponym output.