English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Extracting precise geographical information from textual contents is crucialin a plethora of applications. For example, during hazardous events, a robustand unbiased toponym extraction framework can provide an avenue to tie thelocation concerned to the topic discussed by news media posts and pinpointhumanitarian help requests or damage reports from social media. Early studieshave leveraged rule-based, gazetteer-based, deep learning, and hybridapproaches to address this problem. However, the performance of existing toolsis deficient in supporting operations like emergency rescue, which relies onfine-grained, accurate geographic information. The emerging pretrained languagemodels can better capture the underlying characteristics of text information,including place names, offering a promising pathway to optimize toponymrecognition to underpin practical applications. In this paper, TopoBERT, atoponym recognition module based on a one dimensional Convolutional NeuralNetwork (CNN1D) and Bidirectional Encoder Representation from Transformers(BERT), is proposed and fine-tuned. Three datasets (CoNLL2003-Train,Wikipedia3000, WNUT2017) are leveraged to tune the hyperparameters, discoverthe best training strategy, and train the model. Another two datasets(CoNLL2003-Test and Harvey2017) are used to evaluate the performance. Threedistinguished classifiers, linear, multi-layer perceptron, and CNN1D, arebenchmarked to determine the optimal model architecture. TopoBERT achievesstate-of-the-art performance (f1-score=0.865) compared to the other fivebaseline models and can be applied to diverse toponym recognition tasks withoutadditional training.

TopoBERT: Plug and Play Toponym Recognition Module Harnessing Fine-tuned BERT