Analysis of Issues on COVID-19 Blues Using Big data
Title
Analysis of Issues on COVID-19 Blues Using Big data
Alternative Author(s)
Kim, Yeon-Jung
Keyword
COVID-19 Blues
; Big Data
; Text Mining
Publication Year
2023-06-30
Publisher
Korea Institute for Health and Social Affairs
Citation
Health and Social Welfare Review Vol.43 No.2, pp.29-47
Abstract
The purpose of this study was to analyze big data related to the “COVID-19 blues” to explore major issues and public response. The data were collected using web crawling from January 2020 through March 2021 and analyzed using text mining analysis. The results indicated that the main keywords related to Corona blues were ‘overcoming,’ ‘anxiety,’ ‘person,’ ‘mind,’ ‘prolonged,’ ‘symptom,’ ‘stress,’ and ‘sequela.’ The main N-gram keywords in the data were depression-anxiety, depression-overcoming, and corona-sequela. In relation to COVID-19 blues, words with high co-occurrence frequencies were anxiety, blue, and depression. Our keyword correlation analysis found that the words most related to ‘corona’ were ‘blue’, ‘symptoms’, and ‘restoration’, and the words most related to ‘blues’ were ‘depression’, ‘friend’, and ‘thought’. Topic modeling indicated that 10 topics were ‘confusion due to working from home’, ‘nearby mental health services’, ‘symptoms of COVID-19 blues’, ‘transition to depression’, ‘withdrawal in interpersonal relationships’, ‘youth academic turmoil’, ‘psychological difficulties of youth’, ‘the aftereffects of COVID-19’, ‘birthday blues’, and ‘weight gain’. Big data analysis revealed the need for more inclusive strategies of mental support for COVID-19 blues. Implications for intervention and recommendations for further research are provided.