Xuemin Yu

Xuemin Yu

Auto-generated profile, based on publicly available information

Xuemin Yu

Faculty of Computer Science, Dalhousie University, Halifax, Nova Scotia, Canada

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 September 2021 in IEEE Access
Verified
Journal Article
Open access
XYXuemin Yu
Xuemin Yu
MFMartha Dais Ferreira
Martha Dais Ferreira
Fernando PaulovichFernando Paulovich
Fernando Paulovich
As governments take measures against the COVID-19, the epidemic situation is expected to improve. Consequently, public sentiment is likely to fluctuate during this change, influencing the best course of action. The detection of public sentiment can help the service of various industries and the government’s subsequent formulation of measures and policies. In recent times, social media has become a prevalent way for the public to express emotions and opinions. Therefore, sentiment analysis of social media texts may detect and provide a valuable indication of public sentiment. We present Senti-COVID19, an interactive visual analytic system for reflecting and analyzing public sentiment and detecting sentiment fluctuation triggers on social media. Compared to other visual analytic tools, Senti-COVID19 focuses on sentiment analysis of COVID-19 related tweets and classifies them based on the dominant sentiment category. The system emphasizes analyzing tweets that mainly discuss the epidemic situation with keywords and hashtags related to COVID-19. Senti-COVID19 adopts lexicon-based sentiment analysis to divulge the public sentiment to COVID-19 events, employing libraries to extract keywords and statistics for providing detailed information. In addition, it offers visualization views for presenting the analysis results and extracted information.We design evaluations and a usage scenario, which we used to assess the effectiveness of Senti-COVID19. The results show that Senti-COVID19 can effectively analyze sentiment from social media text and provide interactive visual analysis for users to explore derived data and identify insights from the front-end.