Covid-19 vaccines in Italian public opinion: Identifying key issues using Twitter and Natural Language Processing

Abstract

The COVID-19 pandemic has changed society and people’s lives. The vaccination campaign started December 27th 2020 in Italy, together with most countries in the European Union. Social media platforms can offer relevant information about how citizens have experienced and perceived the availability of vaccines and the start of the vaccination campaign. This study aims to use machine learning methods to extract sentiments and topics relating to COVID-19 vaccination from Twitter. Between February and May 2021, we collected over 71,000 tweets containing vaccines-related keywords from Italian Twitter users. To get the dominant sentiment throughout the Italian population, spatial and temporal sentiment analysis was performed using VADER, highlighting sentiment fluctuations strongly influenced by news of vaccines’ side effects. Additionally, we investigated the opinions of Italians with respect to different vaccine brands. As a result, ‘Oxford-AstraZeneca’ vaccine was the least appreciated among people. The application of the Dynamic Latent Dirichlet Allocation (DLDA) model revealed three fundamental topics, which remained stable over time: vaccination plan info, usefulness of vaccinating and concerns about vaccines (risks, side effects and safety). To the best of our current knowledge, this one the first study on Twitter to identify opinions about COVID-19 vaccination in Italy and their progression over the first months of the vaccination campaign. Our results can help policymakers and research communities track public attitudes towards COVID-19 vaccines and help them make decisions to promote the vaccination campaign.

A natural language processing approach for analyzing COVID-19 vaccination response in multi-language and geo-localized tweets

“…To the best of our knowledge, the majority of previous works on opinions toward COVID-19 vaccination on social media have focused on a single location and a single language [9] , [11] , [12] . On the other hand, our study collects posts regardless of location and language.…”

Lexicon-based sentiment analysis to detect opinions and attitude towards COVID-19 vaccines on Twitter in Italy

“…The aim of Stracqualursi and Agati [24] was to assess public opinion and perception on COVID-19 vaccines in Italy using 71,000 tweets containing vaccines-related keywords from Italian Twitter users, over the period February 1st to May 31st 2021. To determine the prevalent sentiment, spatial and temporal sentiment analysis was performed using VADER, and findings showed that sentiment fluctuations were highly influenced by news of vaccines’ side effects.…”

Year 2022 in Medical Natural Language Processing: Availability of Language Models as a Step in the Democratization of NLP in the Biomedical Area

“…Nevertheless, the trends observed within this set of keywords are also reflected in the analysis provided in the following sections. [23], construction of cohorts of similar patients [24], processing of electronic medical records [25], understanding of patient's answers in a French medical chatbot [26]; • German: evaluation of Transformers on clinical notes [27]; • Greek: improving the performance of localized healthcare virtual assistants [28]; • Hindi: classification of COVID-19 texts [29], chatbot for information sexual and reproductive health for young people [30]; • Italian: analysis of social media for quality of life in Parkinson's patients [31], sentiment analysis of opinion on COVID-19 vaccines [32,33], estimation of the incidence of infectious disease cases [34]; • Japanese: understanding psychiatric illness [35], detection of adverse events from narrative clinical documents [36]; • Korean: BERT model for processing med-ical documents [37], sentiment analysis of tweets about COVID-19 vaccines [38];…”