Background
The COVID-19 pandemic brought unforeseen challenges that could forever change the way societies prioritize and deal with public health issues. The approaches to contain the spread of the virus have entailed governments issuing recommendations on social distancing, lockdowns to restrict movements, and suspension of services.
Objective
There are concerns that the COVID-19 crisis and the measures adopted by countries in response to the pandemic may have led to an upsurge in violence against children. Added stressors placed on caregivers, economic uncertainty, job loss or disruption to livelihoods and social isolation, may have led to a rise in children’s experience of violence in the home. Extended online presence by children may have resulted in increased exposure to abusive content and cyberbullying.
Participants and setting
This study uses testimonial-based and conversational-based data collected from social media users.
Methods
Conversations on Twitter were reviewed to measure increases in abusive or hateful content, and cyberbullying, while testimonials from Reddit forums were examined to monitor changes in references to family violence before and after the start of the stayat-home restrictions.
Results
Violence-related subreddits were among the topics with the highest growth after the COVID-19 outbreak. The analysis of Twitter data shows a significant increase in abusive content generated during the stay-at-home restrictions.
Conclusions
The collective experience of the COVID-19 pandemic and related containment measures offers insights into the wide-ranging risks that children are exposed to in times of crisis. As societies shift towards a new normal, which places emerging technology, remote working and online learning at its center, and in anticipation of similar future threats, governments and other stakeholders need to put in place measures to protect children from violence.
Disasters strike communities around the world, with a reduced time-frame for warning and action leaving behind high rates of damage, mortality, and years in rebuilding efforts. For the past decade, social media has indicated a positive role in communicating before, during, and after disasters. One important question that remained un-investigated is that whether social media efficiently connect affected individuals to disaster relief agencies, and if not, how AI models can use historical data from previous disasters to facilitate information exchange between the two groups. In this study, the BERT model is first fine-tuned using historical data and then it is used to classify the tweets associated with hurricanes Dorian and Harvey based on the type of information provided; and alongside, the network between users is constructed based on the retweets and replies on Twitter. Afterwards, some network metrics are used to measure the diffusion rate of each type of disaster-motivated information. The results show that the messages by disaster eyewitnesses get the least spread while the posts by governments and media have the highest diffusion rates through the network. Additionally, the “cautions and advice” messages get the most spread among other information types while “infrastructure and utilities” and “affected individuals” messages get the least diffusion even compared with “sympathy and support”. The analysis suggests that facilitating the propagation of information provided by affected individuals, using AI models, will be a valuable strategy to pursue in order to accelerate communication between affected individuals and survival groups during the disaster and aftermath.
This paper presents the deep-learning model that is submitted to the SemEval-2020 Task 5 competition: "Detecting Counterfactuals". We participated in both Subtask1 and Subtask2. The model proposed in this paper ranked 2nd in Subtask2: "Detecting antecedent and consequence". Our model approaches the task as a sequence labeling. The architecture is built on top of BERT; and a multi-head attention layer with label masking is used to benefit from the mutual information between nearby labels. Also, for prediction, a multi-stage algorithm is used in which the model finalize some predictions with higher certainty in each step and use them in the following. Our results show that masking the labels not only is an efficient regularization method but also improves the accuracy of the model compared with other alternatives like CRF. Label masking can be used as a regularization method in sequence labeling. Also, it improves the performance of the model by learning the specific patterns in the target variable.
The COVID‐19 pandemic presented many challenges, one of them being the imposition of “work‐at‐home” policies in March 2020. The Systems Engineering Research Center (SERC) and the International Council on Systems Engineering (INCOSE) conducted two online surveys—one during the first months of the pandemic in 2020 and the second survey 1 year after, in March 2021—to understand the impact of these policies within the systems engineering community. The surveys' format consisted of multiple‐choice questions and open‐answer questions, which were analyzed using LDA for topic modeling. Data were also collected from social media during the same timeframes to compare the feelings and experiences of systems engineers with those of the general population.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.