학술논문

Xenophobic Events vs. Refugee Population -- Using GDELT to Identify Countries with Disproportionate Coverage
Document Type
Working Paper
Source
Subject
Computer Science - Computers and Society
Language
Abstract
In this preliminary study, we used the Global Database of Events, Language, and Tone (GDELT) database to examine xenophobic events reported in the media during 2022. We collected a dataset of 2,778 unique events and created a choropleth map illustrating the frequency of events scaled by the refugee population's proportion in each host country. We identified the top 10 countries with the highest scaled event frequencies among those with more than 50,000 refugees. Contrary to the belief that hosting a significant number of forced migrants results in higher xenophobic incidents, our findings indicate a potential connection to political factors. We also categorized the 20 root event codes in the CAMEO event data as either "Direct" or "Indirect". Almost 90% of the events related to refugees in 2022 were classified as "Indirect".
Comment: 10 pages, 2 figures, accepted as a Working Paper at SBP-BRiMS 2023. arXiv admin note: text overlap with arXiv:2305.01708