2022/12/15 by Abdul Sittar, Sittar, Abdul, Daniela Major +7
Computer Science · Social Sciences · #Computational and Text Analysis Methods #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Sentiment Analysis and Opinion Mining
paper · pdf · doi:10.48550/arxiv.2212.13875
openalex publication_date 2022/12/15 · openalex created_date 2023/01/06 · openalex updated_date 2026/07/28
The purpose of this study is to analyse COVID-19 related news published across different geographical places, in order to gain insights in reporting differences. The COVID-19 pandemic had a major outbreak in January 2020 and was followed by different preventive measures, lockdown, and finally by the process of vaccination. To date, more comprehensive analysis of news related to COVID-19 pandemic are missing, especially those which explain what aspects of this pandemic are being reported by newspapers inserted in different economies and belonging to different political alignments. Since LDA is often less coherent when there are news articles published across the world about an event and you look answers for specific queries. It is because of having semantically different content. To address this challenge, we performed pooling of news articles based on information retrieval using TF-IDF score in a data processing step and topic modeling using LDA with combination of 1 to 6 ngrams. We used VADER sentiment analyzer to analyze the differences in sentiments in news articles reported across different geographical places. The novelty of this study is to look at how COVID-19 pandemic was reported by the media, providing a comparison among countries in different political and economic contexts. Our findings suggest that the news reporting by newspapers with different political alignment support the reported content. Also, economic issues reported by newspapers depend on economy of the place where a newspaper resides.