Citerank: (7) 679750Ali AsgaryAssociate Professor and Associate Director, Advanced Disaster, Emergency and Rapid Response Simulation (ADERSIM) in the School of Administrative Studies, and Adjunct Professor in the School of Information Technology, at York University.10019D3ABAB, 679812Jianhong WuProfessor Jianhong Wu is a University Distinguished Research Professor and Senior Canada Research Chair in industrial and applied mathematics at York University. He is also the NSERC Industrial Research Chair in vaccine mathematics, modelling, and manufacturing. 10019D3ABAB, 679815Jude KongDr. Jude Dzevela Kong is an Assistant Professor in the Department of Mathematics and Statistics at York University and the founding Director of the Africa-Canada Artificial Intelligence and Data Innovation Consortium (ACADIC). 10019D3ABAB, 701037MfPH – Publications144B5ACA0, 703957Economics859FDEF6, 704045Covid-19859FDEF6, 715767Woldegebriel Assefa WoldegerimaDr. Woldegerima, knows as "Assefa", is an Assistant Professor at the Department of Mathematics and Statistics at York University.10019D3ABAB

[PLOS ONE, 24 August 2022]

The COVID-19 pandemic has had a devastating impact on the global economy. In this paper, we use the Phillips curve to compare and analyze the macroeconomics of three different countries with distinct income levels, namely, lower-middle (Nigeria), upper-middle (South Africa), and high (Canada) income. We aim to (1) find macroeconomic changes in the three countries during the pandemic compared to pre-pandemic time, (2) compare the countries in terms of response to the COVID-19 economic crisis, and (3) compare their expected economic reaction to the COVID-19 pandemic in the near future. An advantage to our work is that we analyze macroeconomics on a monthly basis to capture the shocks and rapid changes caused by on and off rounds of lockdowns. We use the volume and social sentiments of the Twitter data to approximate the macroeconomic statistics. We apply four different machine learning algorithms to estimate the unemployment rate of South Africa and Nigeria on monthly basis. The results show that at the beginning of the pandemic the unemployment rate increased for all the three countries. However, Canada was able to control and reduce the unemployment rate during the COVID-19 pandemic. Nonetheless, in line with the Phillips curve short-run, the inflation rate of Canada increased to a level that has never occurred in more than fifteen years. Nigeria and South Africa have not been able to control the unemployment rate and did not return to the pre-COVID-19 level. Yet, the inflation rate has increased in both countries. The inflation rate is still comparable to the pre-COVID-19 level in South Africa, but based on the Phillips curve short-run, it will increase further, if the unemployment rate decreases. Unfortunately, Nigeria is experiencing a horrible stagflation and a wild increase in both unemployment and inflation rates. This shows how vulnerable lower-middle-income countries could be to lockdowns and economic restrictions. In the near future, the main concern for all the countries is the high inflation rate. This work can potentially lead to more targeted and publicly acceptable policies based on social media content.

[Frontiers in Public Health, 2 December 2022]

The global economy has been hard hit by the COVID-19 pandemic. Many countries are experiencing a severe and destructive recession. A significant number of firms and businesses have gone bankrupt or been scaled down, and many individuals have lost their jobs. The main goal of this study is to support policy- and decision-makers with additional and real-time information about the labor market flow using Twitter data. We leverage the data to trace and nowcast the unemployment rate of South Africa during the COVID-19 pandemic. First, we create a dataset of unemployment-related tweets using certain keywords. Principal Component Regression (PCR) is then applied to nowcast the unemployment rate using the gathered tweets and their sentiment scores. Numerical results indicate that the volume of the tweets has a positive correlation, and the sentiments of the tweets have a negative correlation with the unemployment rate during and before the COVID-19 pandemic. Moreover, the now-casted unemployment rate using PCR has an outstanding evaluation result with a low Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), Symmetric MAPE (SMAPE) of 0.921, 0.018, 0.018, respectively and a high R2-score of 0.929.

[Data in Brief, 6 June 2023]

After struggling with COVID-19 pandemic for two years, the world is finally recovering from this crisis. Nonetheless, another virus, Monkeypox, is quickly spreading throughout the world and in non-endemic regions and continents, threatening the world to a new pandemic. Twitter as a popular social media has successfully been used for predicting and controlling outbreaks. Much research previously has been done for building early warning systems, trend prediction, and misinformation and fake news detection. Since tweets are not accessible to all researchers, in this work, a publicly available dataset containing 2400202 tweets gathered from May first to December twenty-fifth, 2022 is presented. Twitter developers academic researcher API which returns all the tweets matching a given query was used to gather the dataset. To this end, the full archive search and keywords related to Monkeypox and its equivalents in other languages, i.e. Monkeypox or “monkey pox” or “viruela dei mono” or “variole du singe” or “variola do macoco” were used. The retweets were excluded using the negation operator, and the tweet ids and user ids were extracted and shared with public. Approximately, 1.79 percent (43047 number) of tweets were geotagged. To visualize the geotagged tweets, the longitude and latitude of the bounding box coordinates were averaged. This work will help researchers shed light on the news, patterns, and on-going discussions of Monkeypox on social media, identify hotspots, and help contain the Monkeypox virus.

[Journal of Medical Internet Research, 1 May 2023]

Background: The global Mpox (formerly, Monkeypox) outbreak is disproportionately affecting the gay and bisexual men having sex with men community.

Objective: The aim of this study is to use social media to study country-level variations in topics and sentiments toward Mpox and Two-Spirit, Lesbian, Gay, Bisexual, Transgender, Queer or Questioning, Intersex, Asexual (2SLGBTQIAP+)–related topics. Previous infectious outbreaks have shown that stigma intensifies an outbreak. This work helps health officials control fear and stop discrimination.

Methods: In total, 125,424 Twitter and Facebook posts related to Mpox and the 2SLGBTQIAP+ community were extracted from May 1 to December 25, 2022, using Twitter application programming interface academic accounts and Facebook-scraper tools. The tweets’ main topics were discovered using Latent Dirichlet Allocation in the sklearn library. The pysentimiento package was used to find the sentiments of English and Spanish posts, and the CamemBERT package was used to recognize the sentiments of French posts. The tweets’ and Facebook posts’ languages were understood using the Twitter application programming interface platform and pycld3 library, respectively. Using ArcGis Online, the hot spots of the geotagged tweets were identified. Mann-Whitney U, ANOVA, and Dunn tests were used to compare the sentiment polarity of different topics and countries.

Results: The number of Mpox posts and the number of posts with Mpox and 2SLGBTQIAP+ keywords were 85% correlated (P<.001). Interestingly, the number of posts with Mpox and 2SLGBTQIAP+ keywords had a higher correlation with the number of Mpox cases (correlation=0.36, P<.001) than the number of posts on Mpox (correlation=0.24, P<.001). Of the 10 topics, 8 were aimed at stigmatizing the 2SLGBTQIAP+ community, 3 of which had a significantly lower sentiment score than other topics (ANOVA P<.001). The Mann-Whitney U test shows that negative sentiments have a lower intensity than neutral and positive sentiments (P<.001) and neutral sentiments have a lower intensity than positive sentiments (P<.001). In addition, English sentiments have a higher negative and lower neutral and positive intensities than Spanish and French sentiments (P<.001), and Spanish sentiments have a higher negative and lower positive intensities than French sentiments (P<.001). The hot spots of the tweets with Mpox and 2SLGBTQIAP+ keywords were recognized as the United States, the United Kingdom, Canada, Spain, Portugal, India, Ireland, and Italy. Canada was identified as having more tweets with negative polarity and a lower sentiment score (P<.04).

Conclusions: The 2SLGBTQIAP+ community is being widely stigmatized for spreading the Mpox virus on social media. This turns the community into a highly vulnerable population, widens the disparities, increases discrimination, and accelerates the spread of the virus. By identifying the hot spots and key topics of the related tweets, this work helps decision makers and health officials inform more targeted policies.

[Frontiers in Digital Health, 28 June 2023]

Background: The use of social media data provides an opportunity to complement traditional influenza and COVID-19 surveillance methods for the detection and control of outbreaks and informing public health interventions.

Objective: The first aim of this study is to investigate the degree to which Twitter users disclose health experiences related to influenza and COVID-19 that could be indicative of recent plausible influenza cases or symptomatic COVID-19 infections. Second, we seek to use the Twitter datasets to train and evaluate the classification performance of Bidirectional Encoder Representations from Transformers (BERT) and variant language models in the context of influenza and COVID-19 infection detection.

Methods: We constructed two Twitter datasets using a keyword-based filtering approach on English-language tweets collected from December 2016 to December 2022 in Saskatchewan, Canada. The influenza-related dataset comprised tweets filtered with influenza-related keywords from December 13, 2016, to March 17, 2018, while the COVID-19 dataset comprised tweets filtered with COVID-19 symptom-related keywords from January 1, 2020, to June 22, 2021. The Twitter datasets were cleaned, and each tweet was annotated by at least two annotators as to whether it suggested recent plausible influenza cases or symptomatic COVID-19 cases. We then assessed the classification performance of pre-trained transformer-based language models, including BERT-base, BERT-large, RoBERTa-base, RoBERT-large, BERTweet-base, BERTweet-covid-base, BERTweet-large, and COVID-Twitter-BERT (CT-BERT) models, on each dataset. To address the notable class imbalance, we experimented with both oversampling and undersampling methods.

Results: The influenza dataset had 1129 out of 6444 (17.5%) tweets annotated as suggesting recent plausible influenza cases. The COVID-19 dataset had 924 out of 11939 (7.7%) tweets annotated as inferring recent plausible COVID-19 cases. When compared against other language models on the COVID-19 dataset, CT-BERT performed the best, supporting the highest scores for recall (94.8%), F1(94.4%), and accuracy (94.6%). For the influenza dataset, BERTweet models exhibited better performance. Our results also showed that applying data balancing techniques such as oversampling or undersampling method did not lead to improved model performance.

Conclusions: Utilizing domain-specific language models for monitoring users’ health experiences related to influenza and COVID-19 on social media shows improved classification performance and has the potential to supplement real-time disease surveillance.

[JMIR Infodemiology, 31 May 2023]

Background: Alternative cancer treatment is associated with a greater risk of death than cancer patients undergoing conventional treatments. Anecdotal evidence suggests cancer patients view paid advertisements promoting alternative cancer treatment on social media, but the extent and nature of this advertising remain unknown. This context suggests an urgent need to investigate alternative cancer treatment advertising on social media.

Objective: This study aimed to systematically analyze the advertising activities of prominent alternative cancer treatment practitioners on Meta platforms, including Facebook, Instagram, Messenger, and Audience Network. We specifically sought to determine (1) whether paid advertising for alternative cancer treatment occurs on Meta social media platforms, (2) the strategies and messages of alternative cancer providers to reach and appeal to prospective patients, and (3) how the efficacy of alternative treatments is portrayed.

Methods: Between December 6, 2021, and December 12, 2021, we collected active advertisements from alternative cancer clinics using the Meta Ad Library. The information collected included identification number, URL, active/inactive status, dates launched/ran, advertiser page name, and a screenshot (image) or recording (video) of the advertisement. We then conducted a content analysis to determine how alternative cancer providers communicate the claimed benefits of their services and evaluated how they portrayed alternative cancer treatment efficacy.

Results: We identified 310 paid advertisements from 11 alternative cancer clinics on Meta (Facebook, Instagram, or Messenger) marketing alternative treatment approaches, care, and interventions. Alternative cancer providers appealed to prospective patients through eight strategies: (1) advertiser representation as a legitimate medical provider (n=289, 93.2%); (2) appealing to persons with limited treatments options (n=203, 65.5%); (3) client testimonials (n=168, 54.2%); (4) promoting holistic approaches (n=121, 39%); (5) promoting messages of care (n=81, 26.1%); (6) rhetoric related to science and research (n=72, 23.2%); (7) rhetoric pertaining to the latest technology (n=63, 20.3%); and (8) focusing treatment on cancer origins and cause (n=43, 13.9%). Overall, 25.8% (n=80) of advertisements included a direct statement claiming provider treatment can cure cancer or prolong life.

Conclusions: Our results provide evidence alternative cancer providers are using Meta advertising products to market scientifically unsupported cancer treatments. Advertisements regularly referenced “alternative” and “natural” treatment approaches to cancer. Imagery and text content that emulated evidence-based medical providers created the impression that the offered treatments were effective medical options for cancer. Advertisements exploited the hope of patients with terminal and poor prognoses by sharing testimonials of past patients who allegedly were cured or had their lives prolonged. We recommend that Meta introduce a mandatory, human-led authorization process that is not reliant upon artificial intelligence for medical-related advertisers before giving advertising permissions. Further research should focus on the conflict of interest between social media platforms advertising products and public health.