Spatio-Temporal Distribution of Negative Emotions in New York City After a Natural Disaster as Seen in Social Media

Feasibility of big data analytics in disaster psychiatry: Impact of Seoul Itaewon tragedy on sentiment distribution on Twitter

Numerous studies have highlighted the significant impact of disasters on mental health, often leading to psychiatric disorders among affected individuals. Timely identification of disaster-related mental health problems is crucial to prevent long-term negative consequences and improve individual and community resilience. To address the limitations of prior research that has focused solely on isolated incidents, we analyzed the impact of a recurring Halloween event in Itaewon, South Korea, which culminated tragically in a crowd crush incident in 2022. We conducted sentiment analysis on big data from Korean Twitter to gauge the impact of this disaster on public sentiment. We collected tweets 2 weeks before and after the annual festival from 2020 to 2022, allowing for the consideration of variability across years and days before the disaster. Using a pre-trained RoBERTa neural network model fine-tuned with public sentiment datasets, we categorized tweets into seven pre-defined emotional categories: Anger, sadness, happiness, disgust, fear, surprise, and neutrality. These sentiments were then analyzed as daily time-series data. The overall tweet volume across all sentiment categories increased, particularly showing an increase in the number of tweets indicating "Sadness" in 2022 compared with that in previous years. Post-disaster, a substantial increase was noted in the proportion of tweets expressing "Sadness" and "Fear." This trend was confirmed by Seasonal Autoregressive Integrated Moving Average with Exogenous Regressor models. Notably, there was an increase in the number of tweets expressing all sentiments, including "Happy." However, significant changes in proportions were observed only in tweets categorized as expressing "Sadness" [0.046 (95% CI: 0.024-0.068, P < 0.0001)] and "Fear" [0.033 (95% CI: 0.014-0.051, P < 0.0001)]. Our study demonstrates the feasibility of using sentiment data from social media, combined with sentiment classification, to assess distinct public mental health features following disasters. This approach provides valuable insights into the emotional impact of each event.

Social media insights into spatio-temporal emotional responses to COVID-19 crisis

The Coronavirus pandemic has presented multifaceted challenges in urban emotional well-being and mental health management. Our study presents a spatio-temporal sentiment mining (STSM) framework to address these challenges, focusing on the space-time geography and environmental psychology. This framework analyzes the distribution and trends of 6 categories of public sentiments in Shanghai during the COVID-19 crisis, considering the potential urban spatial influencing factors. The research specifically draws on social media data temporally coinciding with the spread of COVID-19 and the pre-trained language model RoBERTa-wwm-ext to classify public sentiment, in order to characterize the distribution and trends of dominant urban sentiment under the influence of epidemic at different phases. The interactions between urban geospatial features and sentiments are further modelled and explained using LightGBM algorithm and SHapley Additive exPlanations (SHAP) technique. The experimental findings reveal the subtle yet dynamic impact of the urban environment on the long-term spatial variation and trends of public sentiment under the epidemic, with green spaces and socio-economic status emerging as significant factors. Regions with higher permanent population consumption demonstrated more positive sentiments, underscoring the significance of socio-economic factors in urban planning and public health policy. This research offers the most extensive analysis to date on the influence of urban characteristics on public sentiment during Shanghai's epidemic life cycle also lays the groundwork for applying the STSM framework in future crises beyond COVID-19.

The Role of Social Media in Building Pandemic Resilience in an Urban Community: A Qualitative Case Study

This paper explores the influence of social media in fostering resilience within an urban spatial context, specifically in Bangalore, India, during the COVID-19 lockdown, a period marked by a surge in digital communication due to movement restrictions. To control the rapid spread of the virus, over 1.38 billion people were given stay-at-home orders by the government of India during the onset of the pandemic. The restrictions in movement forced individuals to shift to online modes of connection and communication. As the field of digital epidemiology, that is, the use of digital tools and data to understand and improve health took center stage during the pandemic, the focus shifted towards the social media landscape, which is often associated with its negative aspects, such as misinformation. However, this paper delves into social media's potential to build resilience on a local scale, particularly given its increased usage during the pandemic. Through in-depth online interviews with eight urban residents, we conducted a thematic analysis to understand social media's role during the lockdown. Results indicate that social media facilitated effective information exchange and fostered a sense of community. Furthermore, it engendered an environment conducive to prosocial behavior, a known resilience amplifier. We also highlight the importance of baseline context regarding the users directly engaged in social media data generation with respect to digital epidemiology analytics tools for large-scale social media data and the need for qualitative input feeding into their design. Our study highlights the need for a balanced perspective on social media use in times of crisis, recognizing its potential to boost community resilience in an urban setting, and further enriching digital epidemiology approaches.

Modelling and Analyzing the Semantic Evolution of Social Media User Behaviors during Disaster Events: A Case Study of COVID-19

Public behavior in cyberspace is extremely sensitive to emergency disaster events. Using appropriate methodologies to capture the semantic evolution of social media users’ behaviors and discover how it varies across geographic space and time still presents a significant challenge. This study proposes a novel framework based on complex network, topic model, and GIS to describe the topic change of social media users’ behaviors during disaster events. The framework employs topic modeling to extract topics from social media texts, builds a user semantic evolution model based on a complex network to describe topic dynamics, and analyzes the spatio-temporal characteristics of public semantics evolution. The proposed framework has demonstrated its effectiveness in analyzing the semantic spatial–temporal evolution of Chinese Weibo user behavior during COVID-19. The semantic change in response to COVID-19 was characterized by obvious expansion, frequent change, and gradual stabilization over time. In this case, there were obvious geographical differences in users’ semantic changes, which were mainly concentrated in the capital and economically developed areas. The semantics of users finally focused on specific topics related to positivity, epidemic prevention, and factual comments. Our work provides new insight into the behavioral response to disasters and provides the basis for data-driven public sector decisions. In emergency situations, this model could improve situational assessment, assist decision makers to better comprehend public opinion, and support analysts in allocating resources of disaster relief appropriately.

Monitoring Ecological Conditions by Remote Sensing and Social Media Data—Sanya City (China) as Case Study

The ecological environment is the basis of human survival and development. Effective methods to monitor the ecological environment are essential for the healthy development of human settlements. At present, methods based on remote sensing images and other basic data have played key roles in ecological environment monitoring, providing support for decision-making on local ecological environment protection. However, these data and methods have obvious limitations. On the one hand, they cannot reflect the feelings of human beings about the ecological environment in which they live. On the other hand, it is difficult to capture more detailed information about the ecological environment. Non-professional observation data represented by social media describe the ecological environment from the perspective of the public, which can be a powerful supplement to traditional data. However, these different data sources have their own characteristics and forms, and it is difficult to achieve efficient integration. Therefore, in this paper, we proposed a framework that comprehensively considers social media, remote sensing, and other data to monitor the ecological environment of a study area. First, the framework extracted the ecological environment-related information contained in social media data, including public sentiment information and topic keyword information, by integrating algorithms such as natural language processing and machine learning. Then, we constructed a social semantic network related to the ecological environment based on the extracted information. We used a remote sensing image and other basic data to analyze the ecological sensitivity in the study area. Finally, based on the keyword with spatial location attribute contained in the social semantic network, we established the link between the constructed network and the results of ecological sensitivity analysis to comprehensively analyze the ecological environment in the study area. The comprehensive analysis results not only reflect the distribution of ecological vulnerability in the study area, but also help identify specific areas worthy of attention and the ecological problems faced by these areas. We used the city of Sanya in China as a case study to verify the effectiveness of the method in this paper.

Methods and Applications of Social Media Monitoring of Mental Health During Disasters: Scoping Review

BACKGROUND

With the increasing frequency and magnitude of disasters internationally, there is growing research and clinical interest in the application of social media sites for disaster mental health surveillance. However, important questions remain regarding the extent to which unstructured social media data can be harnessed for clinically meaningful decision-making.

OBJECTIVE

This comprehensive scoping review synthesizes interdisciplinary literature with a particular focus on research methods and applications.

METHODS

A total of 6 health and computer science databases were searched for studies published before April 20, 2021, resulting in the identification of 47 studies. Included studies were published in peer-reviewed outlets and examined mental health during disasters or crises by using social media data.

RESULTS

Applications across 31 mental health issues were identified, which were grouped into the following three broader themes: estimating mental health burden, planning or evaluating interventions and policies, and knowledge discovery. Mental health assessments were completed by primarily using lexical dictionaries and human annotations. The analyses included a range of supervised and unsupervised machine learning, statistical modeling, and qualitative techniques. The overall reporting quality was poor, with key details such as the total number of users and data features often not being reported. Further, biases in sample selection and related limitations in generalizability were often overlooked.

CONCLUSIONS

The application of social media monitoring has considerable potential for measuring mental health impacts on populations during disasters. Studies have primarily conceptualized mental health in broad terms, such as distress or negative affect, but greater focus is required on validating mental health assessments. There was little evidence for the clinical integration of social media-based disaster mental health monitoring, such as combining surveillance with social media-based interventions or developing and testing real-world disaster management tools. To address issues with study quality, a structured set of reporting guidelines is recommended to improve the methodological quality, replicability, and clinical relevance of future research on the social media monitoring of mental health during disasters.

Ethical issues of collecting, storing, and analyzing geo-referenced tweets for mental health research

Spatial approaches to epidemiological research with big social media data provide tremendous opportunities to study the relationship between the socio-ecological context where these data are generated and health indicators of interest. Such research poses a number of ethical challenges, particularly in relation to issues such as privacy, informed consent, data security, and storage. While these issues have received considerable attention by researchers in relation to research for physical health purposes in the past 10 years, there have been few efforts to consider the ethical challenges of conducting mental health research, particularly with geo-referenced social media data. The aim of this article is to identify strengths and limitations of current recommendations to address the specific ethical issues of geo-referenced tweets for mental health research. We contribute to the ongoing debate on the ethical implications of big data research and also provide recommendations to researchers and stakeholders alike on how to tackle them, with a specific focus on the use of geo-referenced data for mental health research purposes. With increasing awareness of data privacy and confidentiality issues (even for non-spatial social media data) it becomes crucial to establish professional standards of conduct so that compliance with ethical standards of conducting research with health-related social media data can be prioritized and easily assessed.

Detecting Suicide Ideation in the Era of Social Media: The Population Neuroscience Perspective

Social media platforms are increasingly used across many population groups not only to communicate and consume information, but also to express symptoms of psychological distress and suicidal thoughts. The detection of suicidal ideation (SI) can contribute to suicide prevention. Twitter data suggesting SI have been associated with negative emotions (e.g., shame, sadness) and a number of geographical and ecological variables (e.g., geographic location, environmental stress). Other important research contributions on SI come from studies in neuroscience. To date, very few research studies have been conducted that combine different disciplines (epidemiology, health geography, neurosciences, psychology, and social media big data science), to build innovative research directions on this topic. This article aims to offer a new interdisciplinary perspective, that is, a Population Neuroscience perspective on SI in order to highlight new ways in which multiple scientific fields interact to successfully investigate emotions and stress in social media to detect SI in the population. We argue that a Population Neuroscience perspective may help to better understand the mechanisms underpinning SI and to promote more effective strategies to prevent suicide timely and at scale.

Emotional processing of sadness and disgust evoked by disaster scenes

OBJECTIVE

Disaster scenes produce long-term negative feelings in those who experience them. Previous studies have focused on mitigating disaster impacts through directed forgetting or conscious suppression. However, the initial emotional processing of disaster scenes is not fully understood, hindering the comprehension of long-term disaster impacts. This study aims to investigate how pictures of disaster scenes evoking disgust and sadness are processed via cortical electrical activity.

METHODS

Pictures of grief and mutilation from disasters were used to evoke sadness and disgust, respectively. Event-related desynchronization (ERD) and event-related potentials (ERPs) were used to quantify the intensity and time-course of emotional processing.

RESULTS

The information processing of emotional pictures was stronger than neutral pictures, represented by greater declines of alpha ERD. In the posterior ERP components of N1 and EPN, amplitudes for emotional pictures were larger than those for neutral pictures, which reflected the effects of arousal on visual perception. In the anterior ERP components of P2, P3, and LPP, disgust pictures showed higher attention attraction and enhanced encoding memory processing.

CONCLUSIONS

Disgust disaster scenarios induced long-term prominent LPP, which may correspond with the long-term negative impacts of the disaster.

Regional Differentiation and Influencing Factor Analysis of Residents' Psychological Status during the Early Stage of the COVID-19 Pandemic in South China

A scientific understanding of the impact of COVID-19 on the psychological status of residents is important for improving medical services and responding to public health emergencies. With the help of some of the most popular network communication tools (including Wechat and Weiboand QQ), online questionnaires were completed by South China citizens during the early stage of the COVID-19 pandemic based on psychological stress theory and using a comprehensive sampling method. Through cooperation with experts from other institutions, the content of the questionnaire was designed to include interviewees’ spatial locations and individual information, identify whether negative emotions were generated, and determine the level of psychological stress and the degree of perception change, etc. According to the data type, mathematical statistics and multiple logistic regression methods were used to examine regional differentiation and influencing factors regarding the psychological stress of residents using 1668 valid questionnaires from 53 municipal administrative units in South China. The results firstly showed that over the whole area there was typical regional differentiation in South China, especially in relation to negative expression and psychological stress, with this feature reflecting the dual urban–rural structure. Secondly, regional differences were obvious. Residents of Hainan showed stronger change of psychological stress than those of the other two provinces. In contrast, Guangdong residents were the least psychological stress, and the concept of a harmonious relationship between human beings and nature was not accepted as well as in the other two provinces. Thirdly, in each province the capital city acted as the regional pole, with greater psychological status. This polarization effect decreased with greater distance, reflecting the theory of growth poles in human geography. Fourthly, gender, education level, occupation, informational correction, and the possibility of infection were notable factors that affected the psychological status of interviewees facing COVID-19. However, the functions were different and were decided by the dependent variable. Lastly, based on conclusions summarized from three perspectives, it was found that regional differentiation, public information, and social structure need to focused upon in order to handle sudden major health issues.

Analysis of social media data for public emotion on the Wuhan lockdown event during the COVID-19 pandemic

BACKGROUND

With outbreaks of COVID-19 around the world, lockdown restrictions are routinely imposed to limit the spread of the virus. During periods of lockdown, social media has become the main channel for citizens to exchange information with others. Public emotions are being generated and shared rapidly online with citizens using internet platforms to reduce anxiety and stress, and stay connected while isolated.

OBJECTIVES

This study aims to explore the regularity of emotional evolution by examining public emotions expressed in online discussions about the Wuhan lockdown event in January 2020.

METHODS

Data related to the Wuhan lockdown was collected from Sina Weibo by web crawler. In this study, the Ortony, Clore, and Collins (OCC) model, Word2Vec, and Bi-directional Long Short-Term Memory model were employed to determine emotional types, train vectorization of words, and identify each text emotion for the training set. Latent Dirichlet Allocation models were also employed to mine the various topic categories, while topic emotional evolution was visualized.

RESULTS

Seven types of emotions and four phases were categorized to describe emotional evolution on the Wuhan lockdown event. The study found that negative emotions such as blame and fear dominated in the early days, and public attitudes towards the lockdown gradually alleviated and reached a balance as the situation improved. Emotional expression about Wuhan lockdown event were significantly related to users' gender, location, and whether or not their account was verified. There were statistically significant correlations between different emotions within the subtle emotional categories. In addition, the evolution of emotions presented a different path due to different topics.

CONCLUSIONS

Multiple emotional categories were determined in our study, providing a detailed and explainable emotion analysis to explored emotional appeal of citizen. The public emotions were gradually easing related to the Wuhan lockdown event, there yet exists regional discrimination and post-traumatic stress disorder in this process, which would lead us to pay continuous attention to citizens lives and psychological status post-pandemic. In addition, this study provided an appropriate method and reference case for the government's public opinion control and emotional appeasement.

Space-Time Dependence of Emotions on Twitter after a Natural Disaster

Natural disasters can have significant consequences for population mental health. Using a digital spatial epidemiologic approach, this study documents emotional changes over space and time in the context of a large-scale disaster. Our aims were to (a) explore the spatial distribution of negative emotional expressions of Twitter users before, during, and after Superstorm Sandy in New York City (NYC) in 2012 and (b) examine potential correlations between socioeconomic status and infrastructural damage with negative emotional expressions across NYC census tracts over time. A total of 984,311 geo-referenced tweets with negative basic emotions (anger, disgust, fear, sadness, shame) were collected and assigned to the census tracts within NYC boroughs between 8 October and 18 November 2012. Global and local univariate and bivariate Moran's I statistics were used to analyze the data. We found local spatial clusters of all negative emotions over all disaster periods. Socioeconomic status and infrastructural damage were predominantly correlated with disgust, fear, and shame post-disaster. We identified spatial clusters of emotional reactions during and in the aftermath of a large-scale disaster that could help provide guidance about where immediate and long-term relief measures are needed the most, if transferred to similar events and on comparable data worldwide.

Association of "#covid19" Versus "#chinesevirus" With Anti-Asian Sentiments on Twitter: March 9-23, 2020

Objectives. To examine the extent to which the phrases, "COVID-19" and "Chinese virus" were associated with anti-Asian sentiments.Methods. Data were collected from Twitter's Application Programming Interface, which included the hashtags "#covid19" or "#chinesevirus." We analyzed tweets from March 9 to 23, 2020, corresponding to the week before and the week after President Donald J. Trump's tweet with the phrase, "Chinese Virus." Our analysis focused on 1 273 141 hashtags.Results. One fifth (19.7%) of the 495 289 hashtags with #covid19 showed anti-Asian sentiment, compared with half (50.4%) of the 777 852 hashtags with #chinesevirus. When comparing the week before March 16, 2020, to the week after, there was a significantly greater increase in anti-Asian hashtags associated with #chinesevirus compared with #covid19 (P < .001).Conclusions. Our data provide new empirical evidence supporting recommendations to use the less-stigmatizing term "COVID-19," instead of "Chinese virus."

Digital Surveillance for Monitoring Environmental Health Threats: A Case Study Capturing Public Opinion from Twitter about the 2019 Chennai Water Crisis

Globally, water scarcity has become a common challenge across many regions. Digital surveillance holds promise for monitoring environmental threats to population health due to severe drought. The 2019 Chennai water crisis in India resulted in severe disruptions to social order and daily life, with local residents suffering due to water shortages. This case study explored public opinion captured through the Twitter social media platform, and whether this information could help local governments with emergency response. Sentiment analysis and topic modeling were used to explore public opinion through Twitter during the 2019 Chennai water crisis. The latent Dirichlet allocation (LDA) method identified topics that were most frequently discussed. A naïve Tweet classification method was built, and Twitter posts (called tweets) were allocated to identified topics. Topics were ranked, and corresponding emotions were calculated. A cross-correlation was performed to examine the relationship between online posts about the water crisis and actual rainfall, determined by precipitation levels. During the Chennai water crisis, Twitter users posted content that appeared to show anxiety about the impact of the drought, and also expressed concerns about the government response. Twitter users also mentioned causes for the drought and potential sustainable solutions, which appeared to be mainly positive in tone. Discussion on Twitter can reflect popular public opinion related to emerging environmental health threats. Twitter posts appear viable for informing crisis management as real-time data can be collected and analyzed. Governments and public health officials should adjust their policies and public communication by leveraging online data sources, which could inform disaster prevention measures.

Space-Time Surveillance of Negative Emotions After Consecutive Terrorist Attacks in London

Terrorist attacks pose significant threats to mental health. There is dearth information about the impact of consecutive terrorist attacks on space-time concentrations of emotional reactions. This study collected (1) Twitter data following the two terrorist attacks in London in March and June of 2017, respectively, and (2) deprivation data at small areal levels in the United Kingdom. The space-time permutation model was used to detect the significant clusters of negative emotions, including fear, sadness, and anger in tweets. Logistic regression models were used to examine the social deprivation of communities associated with negative tweeting. The results reported two significant clusters after the March attack, one was in London, ten days after the attack, and the other was far from the attack site between Manchester and Birmingham, three days after the attack. Attention to the reoccurring attack in June diminished quickly. The socially deprived communities experienced double disadvantage-sending fewer tweets but expressing more negative emotions than their counterparts. The findings suggest that terrorism can affect public emotions far and broad. There is a potential for surveillance to rapidly identify geographically concentrated emotions after consecutive or prolonged disasters using social media data.

Spatio-temporal distribution of negative emotions on Twitter during floods in Chennai, India, in 2015: a post hoc analysis

Background

Natural disasters are known to take their psychological toll immediately, and over the long term, on those living through them. Messages posted on Twitter provide an insight into the state of mind of citizens affected by such disasters and provide useful data on the emotional impact on groups of people. In 2015, Chennai, the capital city of Tamil Nadu state in southern India, experienced unprecedented flooding, which subsequently triggered economic losses and had considerable psychological impact on citizens. The objectives of this study are to (i) mine posts to Twitter to extract negative emotions of those posting tweets before, during and after the floods; (ii) examine the spatial and temporal variations of negative emotions across Chennai city via tweets; and (iii) analyse associations in the posts between the emotions observed before, during and after the disaster.

Methods

Using Twitter’s application programming interface, tweets posted at the time of floods were aggregated for detailed categorisation and analysis. The different emotions were extracted and classified by using the National Research Council emotion lexicon. Both an analysis of variance (ANOVA) and mixed-effect analysis were performed to assess the temporal variations in negative emotion rates. Global and local Moran’s I statistic were used to understand the spatial distribution and clusters of negative emotions across the Chennai region. Spatial regression was used to analyse over time the association in negative emotion rates from the tweets.

Results

In the 5696 tweets analysed around the time of the floods, negative emotions were in evidence 17.02% before, 29.45% during and 11.39% after the floods. The rates of negative emotions showed significant variation between tweets sent before, during and after the disaster. Negative emotions were highest at the time of disaster’s peak and reduced considerably post disaster in all wards of Chennai. Spatial clusters of wards with high negative emotion rates were identified.

Conclusions

Spatial analysis of emotions expressed on Twitter during disasters helps to identify geographic areas with high negative emotions and areas needing immediate emotional support. Analysing emotions temporally provides insight into early identification of mental health issues, and their consequences, for those affected by disasters.

Using Social Media to Mine and Analyze Public Opinion Related to COVID-19 in China

The outbreak of Corona Virus Disease 2019 (COVID-19) is a grave global public health emergency. Nowadays, social media has become the main channel through which the public can obtain information and express their opinions and feelings. This study explored public opinion in the early stages of COVID-19 in China by analyzing Sina-Weibo (a Twitter-like microblogging system in China) texts in terms of space, time, and content. Temporal changes within one-hour intervals and the spatial distribution of COVID-19-related Weibo texts were analyzed. Based on the latent Dirichlet allocation model and the random forest algorithm, a topic extraction and classification model was developed to hierarchically identify seven COVID-19-relevant topics and 13 sub-topics from Weibo texts. The results indicate that the number of Weibo texts varied over time for different topics and sub-topics corresponding with the different developmental stages of the event. The spatial distribution of COVID-19-relevant Weibo was mainly concentrated in Wuhan, Beijing-Tianjin-Hebei, the Yangtze River Delta, the Pearl River Delta, and the Chengdu-Chongqing urban agglomeration. There is a synchronization between frequent daily discussions on Weibo and the trend of the COVID-19 outbreak in the real world. Public response is very sensitive to the epidemic and significant social events, especially in urban agglomerations with convenient transportation and a large population. The timely dissemination and updating of epidemic-related information and the popularization of such information by the government can contribute to stabilizing public sentiments. However, the surge of public demand and the hysteresis of social support demonstrated that the allocation of medical resources was under enormous pressure in the early stage of the epidemic. It is suggested that the government should strengthen the response in terms of public opinion and epidemic prevention and exert control in key epidemic areas, urban agglomerations, and transboundary areas at the province level. In controlling the crisis, accurate response countermeasures should be formulated following public help demands. The findings can help government and emergency agencies to better understand the public opinion and sentiments towards COVID-19, to accelerate emergency responses, and to support post-disaster management.

Factors Associated with Food Insecurity Following Hurricane Harvey in Texas

Food insecurity prevalence among disaster-affected households has been found to be higher than state prevalence in non-disaster times. This study applies a socio-ecological model of post-disaster food insecurity to a nested quota sample (n = 1002) recruited for a web survey from 41 Texas counties affected by Hurricane Harvey 12-15 months post-event. This analysis identifies risk and protective factors for food insecurity. Chi-square analysis was used to examine independent associations between individual, household, and social factors with food insecurity. A multivariate logistic model was fitted and adjusted odds ratios are reported. Economic instability (adjusted odds ratio (OR) 2.43; 95% Confidence Interval (CI) 1.73, 3.41), relocation due to Hurricane Harvey (OR 1.89; CI 1.15, 3.09), major home damage (OR 2.11; CI 1.12, 3.98), non-white race - black (OR 1.79; CI 1.01, 3.18), Hispanic (OR 1.67; CI 1.09, 2.54), other race (OR 4.39; CI 1.96, 9.82) - and community-based organization assistance (1.99; 1.11, 3.58) were risk factors while older age (45-64 years: 0.49; 0.32, 0.73; 65+ years 0.40; 0.22, 0.75), better physical health (0.46; 0.29, 0.71), better mental health (0.46; 0.32, 0.67), and high social support (0.37; 0.25, 0.55) were protective against food insecurity. Disaster policies and programs should address the disproportionate burden on households that relocate or have health conditions. Fostering social support networks, especially among relocated populations, may improve disaster health outcomes.

Digital Innovations for Global Mental Health: Opportunities for Data Science, Task Sharing, and Early Intervention

PURPOSE

Globally, individuals living with mental disorders are more likely to have access to a mobile phone than mental health care. In this commentary, we highlight opportunities for expanding access to and use of digital technologies to advance research and intervention in mental health, with emphasis on the potential impact in lower resource settings.

RECENT FINDINGS

Drawing from empirical evidence, largely from higher income settings, we considered three emerging areas where digital technology will potentially play a prominent role: supporting methods in data science to further our understanding of mental health and inform interventions, task sharing for building workforce capacity by training and supervising non-specialist health workers, and facilitating new opportunities for early intervention for young people in lower resource settings. Challenges were identified related to inequities in access, threats of bias in big data analyses, risks to users, and need for user involvement to support engagement and sustained use of digital interventions.

SUMMARY

For digital technology to achieve its potential to transform the ways we detect, treat, and prevent mental disorders, there is a clear need for continued research involving multiple stakeholders, and rigorous studies showing that these technologies can successfully drive measurable improvements in mental health outcomes.

A Systematic Review of Techniques Employed for Determining Mental Health Using Social Media in Psychological Surveillance During Disasters

During disasters, people share their thoughts and emotions on social media and also provide information about the event. Mining the social media messages and updates can be helpful in understanding the emotional state of people during such unforeseen events as they are real-time data. The objective of this review is to explore the feasibility of using social media data for mental health surveillance as well as the techniques used for determining mental health using social media data during disasters. PubMed, PsycINFO, and PsycARTICLES databases were searched from 2009 to November 2018 for primary research studies. After screening and analyzing the records, 18 studies were included in this review. Twitter was the widely researched social media platform for understanding the mental health of people during a disaster. Psychological surveillance was done by identifying the sentiments expressed by people or the emotions they displayed in their social media posts. Classification of sentiments and emotions were done using lexicon-based or machine learning methods. It is not possible to conclude that a particular technique is the best performing one, because the performance of any method depends upon factors such as the disaster size, the volume of data, disaster setting, and the disaster web environment.

Using Twitter Data to Monitor Natural Disaster Social Dynamics: A Recurrent Neural Network Approach with Word Embeddings and Kernel Density Estimation

In recent years, Online Social Networks (OSNs) have received a great deal of attention for their potential use in the spatial and temporal modeling of events owing to the information that can be extracted from these platforms. Within this context, one of the most latent applications is the monitoring of natural disasters. Vital information posted by OSN users can contribute to relief efforts during and after a catastrophe. Although it is possible to retrieve data from OSNs using embedded geographic information provided by GPS systems, this feature is disabled by default in most cases. An alternative solution is to geoparse specific locations using language models based on Named Entity Recognition (NER) techniques. In this work, a sensor that uses Twitter is proposed to monitor natural disasters. The approach is intended to sense data by detecting toponyms (named places written within the text) in tweets with event-related information, e.g., a collapsed building on a specific avenue or the location at which a person was last seen. The proposed approach is carried out by transforming tokenized tweets into word embeddings: a rich linguistic and contextual vector representation of textual corpora. Pre-labeled word embeddings are employed to train a Recurrent Neural Network variant, known as a Bidirectional Long Short-Term Memory (biLSTM) network, that is capable of dealing with sequential data by analyzing information in both directions of a word (past and future entries). Moreover, a Conditional Random Field (CRF) output layer, which aims to maximize the transition from one NER tag to another, is used to increase the classification accuracy. The resulting labeled words are joined to coherently form a toponym, which is geocoded and scored by a Kernel Density Estimation function. At the end of the process, the scored data are presented graphically to depict areas in which the majority of tweets reporting topics related to a natural disaster are concentrated. A case study on Mexico's 2017 Earthquake is presented, and the data extracted during and after the event are reported.