Mining Geotagged Tweets: Tracking Spatiotemporal Variation of Mental Health in Canada during COVID-19 Pandemic
dc.contributor.author | Pan, Charlotte Zhuoran | |
dc.date.accessioned | 2023-05-29T17:15:11Z | |
dc.date.available | 2023-05-29T17:15:11Z | |
dc.date.issued | 2023-05-29 | |
dc.date.submitted | 2023-05-21 | |
dc.description.abstract | This thesis explores and analyzes the evolution of the pandemic as a stressor on mental health in Canada through monitoring the sentiment polarity dynamics, emotion trends, and changes in keywords being discussed on Twitter, spanning between January 2020 to December 2022. Leveraging the surging amount of geotagged social media data, this study deploys a combination of machine learning, geospatial mapping, and social sensing as a new approach to observe, quantify and evaluate the evolution of national-wide emotion trends and psychological status along the COVID-19 pandemic timeline in Canada, interpret the underlying key factors and events, and thus inform us on how to mentally “re-start” in the post-pandemic era. The proposed methods include social sensing, large-scale sentiment polarity detection, emotion classification, keyword analysis, and kernel density mapping. The dataset after processing is consisting of 430,399 geo-tagged tweets discussing pandemic subjects posted by Canadian users from January 1, 2020 to December 31, 2022. The results of this study reveal that the overall sentiment and emotion composition was the most optimistic during the early half of the pandemic, from the early spring of 2020 to the summer of 2021, and turned to decline from then to the end of 2022, sending a warning signal in public mental well-being. Beneath this trend, several driving events emerged, ranging from the declaration of state of emergency in March 2020, the peak of vaccine hesitancy in November 2020, the release of new vaccine mandate in January 2022 to the Freedom Convey lasting from January 2022 to February 2022. The results also indicate that there is an observable geospatial disparity in the shifting patterns and the overall mental health levels between Montréal, a French-dominant region, and Vancouver, Calgary, Edmonton, Toronto, and Ottawa-Gatineau, which are English-dominant or bilingual regions. Also, along with a delayed period of peaks and bottoms in sentiment polarity, Toronto is displaying a slightly different mood than the other English-speaking cities. Last but not the least, we propose two action strategies, promoting education on the importance of vaccine behaviours and rebalancing the COVID-19 restrictions, for boosting public confidence regarding the pandemic and rebuilding psychological resilience in the current post-pandemic era. As the first work tracking the long-term mental health of Canada as a country during the pandemic, this study evidences the conclusion that as the global economy starts to recover and the number of cases becomes gradually under control with the availability of the vaccine, the public psychological condition is not lifting as fast as the economy and the physical health in today’s post-pandemic world. | en |
dc.identifier.uri | http://hdl.handle.net/10012/19502 | |
dc.language.iso | en | en |
dc.pending | false | |
dc.publisher | University of Waterloo | en |
dc.subject | geotagged tweets | en |
dc.subject | mental health | en |
dc.subject | Canada | en |
dc.title | Mining Geotagged Tweets: Tracking Spatiotemporal Variation of Mental Health in Canada during COVID-19 Pandemic | en |
dc.type | Master Thesis | en |
uws-etd.degree | Master of Science | en |
uws-etd.degree.department | Geography and Environmental Management | en |
uws-etd.degree.discipline | Geography | en |
uws-etd.degree.grantor | University of Waterloo | en |
uws-etd.embargo.terms | 0 | en |
uws.contributor.advisor | Li, Jonathan | |
uws.contributor.affiliation1 | Faculty of Arts | en |
uws.peerReviewStatus | Unreviewed | en |
uws.published.city | Waterloo | en |
uws.published.country | Canada | en |
uws.published.province | Ontario | en |
uws.scholarLevel | Graduate | en |
uws.typeOfResource | Text | en |