Open AccessASSESSMENT OF ATTITUDE TO ANTI-EPIDEMIC MEASURES ON THE BASIS OF ANALYSIS OF SOCIAL NETWORK CONTENT
Publication year - 2020
Publication title -
vìsnik harkìvsʹkogo nacìonalʹnogo unìversitetu ìmenì v.n. karazìna. serìâ ekonomìka
Language(s) - English
Resource type - Journals
ISSN - 2311-2379
DOI - 10.26565/2311-2379-2020-99-03
Subject(s) - sadness , anger , government (linguistics) , pandemic , pessimism , laughter , psychology , sentiment analysis , social psychology , political science , public relations , covid-19 , computer science , linguistics , artificial intelligence , medicine , philosophy , disease , epistemology , pathology , infectious disease (medical specialty)
The article analyzes the dynamics of the COVID-19 pandemic, measures to counter its spread in the world and Ukraine, and also considers the economic consequences. The concept of trust and its impact on the economy are considered in detail, and indicators of trust in state and local authorities in a pandemic are analyzed. The points of view of users of social networks on the economic consequences of the pandemic are determined. The sample of publications was collected for the April-May 2020 period using 360 unique searches at the crossroads of coronavirus and government topics, including 6,726 posts from Ukrainian Facebook users. The words used in the resulting corpus of texts turned out to be: coronavirus, epidemic, quarantine, mask, government, state, president. The semantic analysis of the corpus, carried out using the Word2Vec toolkit, showed that the posts about coronavirus often discuss the state budget, measures to combat the epidemic and the incidence rate, in connection with quarantine - fines and violations, infographics on anti-epidemiological measures. To analyze user sentiment, dictionaries of positive and negative words were built and analyzed, comparing which, it can be noted that, on average, words with an optimistic tone are used 30% more often than with pessimistic ones. Analysis of the reaction to publications by number and type showed that the word "coronavirus" evokes very contradictory emotions, "laughter" and "anger" are practically on the same level. At the same time, the mention of the words "government" and "quarantine" most often causes "anger" and "sadness", "president" and "economy" - "laughter" and "anger" ("contempt" and "aggression" according to Plutchik's methodology). The article proposes a method for assessing attitudes towards anti-epidemiological measures based on the analysis of the content of social networks, including: 1) collection of data on a selected topic from the Facebook network, 2) initial training and statistical analysis of the text. corpus, 3) semantic analysis of the text corpus, 4) analysis of user sentiments. The assessment obtained using the proposed method is confirmed by the results of a survey in support of the government's work to counter the spread of coronavirus, according to which only about 10% of respondents speak positively about its actions, more than 60% - negatively.Thus, the method for assessing attitudes towards anti-epidemiological measures based on the analysis of the content of social networks is implemented as a set of SQL and Python scripts. This method can be used to regularly monitor public opinion regarding the assessment of work to counter the spread of coronavirus.