Geolocating Political Events in Text

Halterman, Andrew

doi:10.18653/v1/w19-2104

Cited by 11 publications

(10 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In future studies, we would like to further investigate the quality of research publications, e.g., by ranking the publications based on their number of citations. In addition, it is acknowledged that the geospatial map might not exactly reflect the study areas, because the geocoding process used in this study might only achieve 83% accuracy (Halterman 2019). Some location names might be geocoded incorrectly or cannot be geocoded.…”

Section: Discussionmentioning

confidence: 96%

Geospatial and Semantic Mapping Platform for Massive COVID-19 Scientific Publication Search

Gong

et al. 2021

J geovis spat anal

View full text Add to dashboard Cite

Thousands of results of scientific publications on the topic of COVID-19 have emerged within a short time, raising significant challenges for scholars to organize the research, or even synthesize the knowledge in a timely or comprehensive manner. To facilitate this, we have developed a geospatial and semantic mapping platform to search and organize these large and unmapped digital collections. The semantic map visualizes research topics based on customized natural language processing algorithms, which helps users to identify their content of specific interest beyond keyword searches in web search engines. The resultant geospatial map extracts all the location names mentioned in the publications, illustrating where COVID-19 studies have been conducted and where neglected study areas might exist. The system has been deployed as a web service for public access.

show abstract

Section: Discussionmentioning

confidence: 96%

Geospatial and Semantic Mapping Platform for Massive COVID-19 Scientific Publication Search

Gong

et al. 2021

J geovis spat anal

View full text Add to dashboard Cite

show abstract

“…New entrants in this area are the Machine-learning Protest Event Data System (MPEDS) (Hanna, 2017) and the Open Event Data Alliance (OEDA) (Halterman et al, 2017). The next generation of machinecoded event data uses advances in natural language processing, distributed computing, and machine translation to code actors and locations, including in non-English languages (Solaimani et al, 2016, Halterman, 2019. The most advanced datasets in this domain, as of this writing, are ICEWS and Temporally Extended, Regular, Reproducible International Event Records (TERRIER) (Liang et al, 2018).…”

Section: Three Methodologiesmentioning

confidence: 99%

Protest Event Data from Geolocated Social Media Content

Steinert-Threlkeld

Joo

2020

Preprint

View full text Add to dashboard Cite

While it is understood that protester identity, violence, and emotions affect the size of protests, these concepts have proved difficult to measure at the protest-day level. Geolocated text and images from social media can improve these measurements. This advance is demonstrated on protests in Venezuela and Chile; it uncovers more protests in Venezuela and generates new measures in both countries. Moreover, the methodology generates daily city-day protest data in 107 countries containing 82.7% of the world’s population and 97.15% of its GDP. These multimodal protest event data complement existing event datasets, though countries’ population and income constrain the reach of any methodology.

show abstract

“…Chung et al [2] developed a rule based system to identify locations even though it is not explicitly mentioned in the text. Halterman et al [8] developed a CNN-LSTM based network to perform the event-location linking task. A context aware algorithm has been developed in [5] with improvements of street-level geotagging as well of geotagging, even if no place has been mentioned.…”

Section: Related Workmentioning

confidence: 99%

EUDETECTOR: Leveraging Language Model to Identify EU-Related News

Rudra

Tran

Shaltev

2021

Companion Proceedings of the Web Conference 2021

View full text Add to dashboard Cite

News media reflects the present state of a country or region to its audiences. Media outlets of a region post different kinds of news for their local and global audiences. In this paper, we focus on Europe (precisely EU) and propose a method to identify news that has an impact on Europe from any aspect such as financial, business, crime, politics, etc. Predicting the location of the news is itself a challenging task. Most of the approaches restrict themselves towards named entities or handcrafted features. In this paper, we try to overcome that limitation i.e., instead of focusing only on the named entities (Europe location, politicians etc.) and some handcrafted rules, we also explore the context of news articles with the help of pre-trained language model BERT. The auto-regressive language model based European news detector shows about 9-19% improvement in terms of F-score over baseline models. Interestingly, we observe that such models automatically capture named entities, their origin, etc; hence, no separate information is required. We also evaluate the role of such entities in the prediction and explore the tokens that BERT really looks at for deciding the news category. Entities such as person, location, organization turn out to be good rationale tokens for the prediction.

show abstract

Geolocating Political Events in Text

Cited by 11 publications

References 30 publications

Geospatial and Semantic Mapping Platform for Massive COVID-19 Scientific Publication Search

Geospatial and Semantic Mapping Platform for Massive COVID-19 Scientific Publication Search

Protest Event Data from Geolocated Social Media Content

EUDETECTOR: Leveraging Language Model to Identify EU-Related News

Contact Info

Product

Resources

About