JMIR infodemiology最新文献

{"title":"Data Exploration and Classification of News Article Reliability: Deep Learning Study.","authors":"Kevin Zhan,&nbsp;Yutong Li,&nbsp;Rafay Osmani,&nbsp;Xiaoyu Wang,&nbsp;Bo Cao","doi":"10.2196/38839","DOIUrl":"https://doi.org/10.2196/38839","url":null,"abstract":"<p><strong>Background: </strong>During the ongoing COVID-19 pandemic, we are being exposed to large amounts of information each day. This \"infodemic\" is defined by the World Health Organization as the mass spread of misleading or false information during a pandemic. This spread of misinformation during the infodemic ultimately leads to misunderstandings of public health orders or direct opposition against public policies. Although there have been efforts to combat misinformation spread, current manual fact-checking methods are insufficient to combat the infodemic.</p><p><strong>Objective: </strong>We propose the use of natural language processing (NLP) and machine learning (ML) techniques to build a model that can be used to identify unreliable news articles online.</p><p><strong>Methods: </strong>First, we preprocessed the ReCOVery data set to obtain 2029 English news articles tagged with COVID-19 keywords from January to May 2020, which are labeled as reliable or unreliable. Data exploration was conducted to determine major differences between reliable and unreliable articles. We built an ensemble deep learning model using the body text, as well as features, such as sentiment, Empath-derived lexical categories, and readability, to classify the reliability.</p><p><strong>Results: </strong>We found that reliable news articles have a higher proportion of neutral sentiment, while unreliable articles have a higher proportion of negative sentiment. Additionally, our analysis demonstrated that reliable articles are easier to read than unreliable articles, in addition to having different lexical categories and keywords. Our new model was evaluated to achieve the following performance metrics: 0.906 area under the curve (AUC), 0.835 specificity, and 0.945 sensitivity. These values are above the baseline performance of the original ReCOVery model.</p><p><strong>Conclusions: </strong>This paper identified novel differences between reliable and unreliable news articles; moreover, the model was trained using state-of-the-art deep learning techniques. We aim to be able to use our findings to help researchers and the public audience more easily identify false information and unreliable media in their everyday lives.</p>","PeriodicalId":73554,"journal":{"name":"JMIR infodemiology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9516811/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"33486461","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Information Sharing Behaviors of Dietitians and Twitter Users in the Nutrition and COVID-19 Infodemic: Content Analysis Study of Tweets.","authors":"Esther Charbonneau, Sehl Mellouli, Arbi Chouikh, Laurie-Jane Couture, Sophie Desroches","doi":"10.2196/38573","DOIUrl":"10.2196/38573","url":null,"abstract":"<p><strong>Background: </strong>The COVID-19 pandemic has generated an infodemic, an overabundance of online and offline information. In this context, accurate information as well as misinformation and disinformation about the links between nutrition and COVID-19 have circulated on Twitter since the onset of the pandemic.</p><p><strong>Objective: </strong>The purpose of this study was to compare tweets on nutrition in times of COVID-19 published by 2 groups, namely, a preidentified group of dietitians and a group of general users of Twitter, in terms of themes, content accuracy, use of behavior change factors, and user engagement, in order to contrast their information sharing behaviors during the pandemic.</p><p><strong>Methods: </strong>Public English-language tweets published between December 31, 2019, and December 31, 2020, by 625 dietitians from Canada and the United States, and Twitter users were collected using hashtags and keywords related to nutrition and COVID-19. After filtration, tweets were coded against an original codebook of themes and the Theoretical Domains Framework (TDF) for identifying behavior change factors, and were compared to reliable nutritional recommendations pertaining to COVID-19. The numbers of likes, replies, and retweets per tweet were also collected to determine user engagement.</p><p><strong>Results: </strong>In total, 2886 tweets (dietitians, n=1417; public, n=1469) were included in the analyses. Differences in frequency between groups were found in 11 out of 15 themes. Grocery (271/1417, 19.1%), and diets and dietary patterns (n=507, 34.5%) were the most frequently addressed themes by dietitians and the public, respectively. For 9 out of 14 TDF domains, there were differences in the frequency of usage between groups. \"Skills\" was the most used domain by both groups, although they used it in different proportions (dietitians: 612/1417, 43.2% vs public: 529/1469, 36.0%; <i>P</i><.001). A higher proportion of dietitians' tweets were accurate compared with the public's tweets (532/575, 92.5% vs 250/382, 65.5%; <i>P</i><.001). The results for user engagement were mixed. While engagement by likes varied between groups according to the theme, engagement by replies and retweets was similar across themes but varied according to the group.</p><p><strong>Conclusions: </strong>Differences in tweets between groups, notably ones related to content accuracy, themes, and engagement in the form of likes, shed light on potentially useful and relevant elements to include in timely social media interventions aiming at fighting the COVID-19-related infodemic or future infodemics.</p>","PeriodicalId":73554,"journal":{"name":"JMIR infodemiology","volume":null,"pages":null},"PeriodicalIF":3.5,"publicationDate":"2022-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9511036/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40392961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Emotions and Incivility in Vaccine Mandate Discourse: Natural Language Processing Insights.","authors":"Hannah Stevens, Muhammad Ehab Rasul, Yoo Jung Oh","doi":"10.2196/37635","DOIUrl":"10.2196/37635","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Despite vaccine availability, vaccine hesitancy has inhibited public health officials' efforts to mitigate the COVID-19 pandemic in the United States. Although some US elected officials have responded by issuing vaccine mandates, others have amplified vaccine hesitancy by broadcasting messages that minimize vaccine efficacy. The politically polarized nature of COVID-19 information on social media has given rise to incivility, wherein health attitudes often hinge more on political ideology than science.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;To the best of our knowledge, incivility has not been studied in the context of discourse regarding COVID-19 vaccines and mandates. Specifically, there is little focus on the psychological processes that elicit uncivil vaccine discourse and behaviors. Thus, we investigated 3 psychological processes theorized to predict discourse incivility-namely, anxiety, anger, and sadness.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We used 2 different natural language processing approaches: (1) the Linguistic Inquiry and Word Count computational tool and (2) the Google Perspective application programming interface (API) to analyze a data set of 8014 tweets containing terms related to COVID-19 vaccine mandates from September 14, 2021, to October 1, 2021. To collect the tweets, we used the Twitter API Tweet Downloader Tool (version 2). Subsequently, we filtered through a data set of 375,000 vaccine-related tweets using keywords to extract tweets explicitly focused on vaccine mandates. We relied on the Linguistic Inquiry and Word Count computational tool to measure the valence of linguistic anger, sadness, and anxiety in the tweets. To measure dimensions of post incivility, we used the Google Perspective API.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;This study resolved discrepant operationalizations of incivility by introducing incivility as a multifaceted construct and explored the distinct emotional processes underlying 5 dimensions of discourse incivility. The findings revealed that 3 types of emotions-anxiety, anger, and sadness-were uniquely associated with dimensions of incivility (eg, toxicity, severe toxicity, insult, profanity, threat, and identity attacks). Specifically, the results showed that anger was significantly positively associated with all dimensions of incivility (all &lt;i&gt;P&lt;/i&gt;&lt;.001), whereas sadness was significantly positively related to threat (&lt;i&gt;P&lt;/i&gt;=.04). Conversely, anxiety was significantly negatively associated with identity attack (&lt;i&gt;P&lt;/i&gt;=.03) and profanity (&lt;i&gt;P&lt;/i&gt;=.02).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;The results suggest that our multidimensional approach to incivility is a promising alternative to understanding and intervening in the psychological processes underlying uncivil vaccine discourse. Understanding specific emotions that can increase or decrease incivility such as anxiety, anger, and sadness can enable researchers and public health professionals to develop effective inte","PeriodicalId":73554,"journal":{"name":"JMIR infodemiology","volume":null,"pages":null},"PeriodicalIF":3.5,"publicationDate":"2022-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9511016/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9704987","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Negative COVID-19 Vaccine Information on Twitter: Content Analysis.","authors":"Niko Yiannakoulias,&nbsp;J Connor Darlington,&nbsp;Catherine E Slavik,&nbsp;Grant Benjamin","doi":"10.2196/38485","DOIUrl":"https://doi.org/10.2196/38485","url":null,"abstract":"<p><strong>Background: </strong>Social media platforms, such as Facebook, Instagram, Twitter, and YouTube, have a role in spreading anti-vaccine opinion and misinformation. Vaccines have been an important component of managing the COVID-19 pandemic, so content that discourages vaccination is generally seen as a concern to public health. However, not all negative information about vaccines is explicitly anti-vaccine, and some of it may be an important part of open communication between public health experts and the community.</p><p><strong>Objective: </strong>This research aimed to determine the frequency of negative COVID-19 vaccine information on Twitter in the first 4 months of 2021.</p><p><strong>Methods: </strong>We manually coded 7306 tweets sampled from a large sampling frame of tweets related to COVID-19 and vaccination collected in early 2021. We also coded the geographic location and mentions of specific vaccine producers. We compared the prevalence of anti-vaccine and negative vaccine information over time by author type, geography (United States, United Kingdom, and Canada), and vaccine developer.</p><p><strong>Results: </strong>We found that 1.8% (131/7306) of tweets were anti-vaccine, but 21% (1533/7306) contained negative vaccine information. The media and government were common sources of negative vaccine information but not anti-vaccine content. Twitter users from the United States generated the plurality of negative vaccine information; however, Twitter users in the United Kingdom were more likely to generate negative vaccine information. Negative vaccine information related to the Oxford/AstraZeneca vaccine was the most common, particularly in March and April 2021.</p><p><strong>Conclusions: </strong>Overall, the volume of explicit anti-vaccine content on Twitter was small, but negative vaccine information was relatively common and authored by a breadth of Twitter users (including government, medical, and media sources). Negative vaccine information should be distinguished from anti-vaccine content, and its presence on social media could be promoted as evidence of an effective communication system that is honest about the potential negative effects of vaccines while promoting the overall health benefits. However, this content could still contribute to vaccine hesitancy if it is not properly contextualized.</p>","PeriodicalId":73554,"journal":{"name":"JMIR infodemiology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9632001/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40454693","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"COVID-19 Misinformation Detection: Machine-Learned Solutions to the Infodemic.","authors":"Nikhil Kolluri, Yunong Liu, Dhiraj Murthy","doi":"10.2196/38756","DOIUrl":"10.2196/38756","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;The volume of COVID-19-related misinformation has long exceeded the resources available to fact checkers to effectively mitigate its ill effects. Automated and web-based approaches can provide effective deterrents to online misinformation. Machine learning-based methods have achieved robust performance on text classification tasks, including potentially low-quality-news credibility assessment. Despite the progress of initial, rapid interventions, the enormity of COVID-19-related misinformation continues to overwhelm fact checkers. Therefore, improvement in automated and machine-learned methods for an infodemic response is urgently needed.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;The aim of this study was to achieve improvement in automated and machine-learned methods for an infodemic response.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We evaluated three strategies for training a machine-learning model to determine the highest model performance: (1) COVID-19-related fact-checked data only, (2) general fact-checked data only, and (3) combined COVID-19 and general fact-checked data. We created two COVID-19-related misinformation data sets from fact-checked \"false\" content combined with programmatically retrieved \"true\" content. The first set contained ~7000 entries from July to August 2020, and the second contained ~31,000 entries from January 2020 to June 2022. We crowdsourced 31,441 votes to human label the first data set.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The models achieved an accuracy of 96.55% and 94.56% on the first and second external validation data set, respectively. Our best-performing model was developed using COVID-19-specific content. We were able to successfully develop combined models that outperformed human votes of misinformation. Specifically, when we blended our model predictions with human votes, the highest accuracy we achieved on the first external validation data set was 99.1%. When we considered outputs where the machine-learning model agreed with human votes, we achieved accuracies up to 98.59% on the first validation data set. This outperformed human votes alone with an accuracy of only 73%.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;External validation accuracies of 96.55% and 94.56% are evidence that machine learning can produce superior results for the difficult task of classifying the veracity of COVID-19 content. Pretrained language models performed best when fine-tuned on a topic-specific data set, while other models achieved their best accuracy when fine-tuned on a combination of topic-specific and general-topic data sets. Crucially, our study found that blended models, trained/fine-tuned on general-topic content with crowdsourced data, improved our models' accuracies up to 99.7%. The successful use of crowdsourced data can increase the accuracy of models in situations when expert-labeled data are scarce. The 98.59% accuracy on a \"high-confidence\" subsection comprised of machine-learned and human labels sugges","PeriodicalId":73554,"journal":{"name":"JMIR infodemiology","volume":null,"pages":null},"PeriodicalIF":3.5,"publicationDate":"2022-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9987189/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9733183","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Promoting Social Distancing and COVID-19 Vaccine Intentions to Mothers: Randomized Comparison of Information Sources in Social Media Messages.","authors":"David Buller, Barbara Walkosz, Kimberly Henry, W Gill Woodall, Sherry Pagoto, Julia Berteletti, Alishia Kinsey, Joseph Divito, Katie Baker, Joel Hillhouse","doi":"10.2196/36210","DOIUrl":"10.2196/36210","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Social media disseminated information and spread misinformation during the COVID-19 pandemic that affected prevention measures, including social distancing and vaccine acceptance.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;In this study, we aimed to test the effect of a series of social media posts promoting COVID-19 nonpharmaceutical interventions (NPIs) and vaccine intentions and compare effects among 3 common types of information sources: government agency, near-peer parents, and news media.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;A sample of mothers of teen daughters (N=303) recruited from a prior trial were enrolled in a 3 (information source) × 4 (assessment period) randomized factorial trial from January to March 2021 to evaluate the effects of information sources in a social media campaign addressing NPIs (ie, social distancing), COVID-19 vaccinations, media literacy, and mother-daughter communication about COVID-19. Mothers received 1 social media post per day in 3 randomly assigned Facebook private groups, Monday-Friday, covering all 4 topics each week, plus 1 additional post on a positive nonpandemic topic to promote engagement. Posts in the 3 groups had the same messages but differed by links to information from government agencies, near-peer parents, or news media in the post. Mothers reported on social distancing behavior and COVID-19 vaccine intentions for self and daughter, theoretic mediators, and covariates in baseline and 3-, 6-, and 9-week postrandomization assessments. Views, reactions, and comments related to each post were counted to measure engagement with the messages.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Nearly all mothers (n=298, 98.3%) remained in the Facebook private groups throughout the 9-week trial period, and follow-up rates were high (n=276, 91.1%, completed the 3-week posttest; n=273, 90.1%, completed the 6-week posttest; n=275, 90.8%, completed the 9-week posttest; and n=244, 80.5%, completed all assessments). In intent-to-treat analyses, social distancing behavior by mothers (b=-0.10, 95% CI -0.12 to -0.08, &lt;i&gt;P&lt;/i&gt;&lt;.001) and daughters (b=-0.10, 95% CI -0.18 to -0.03, &lt;i&gt;P&lt;/i&gt;&lt;.001) decreased over time but vaccine intentions increased (mothers: b=0.34, 95% CI 0.19-0.49, &lt;i&gt;P&lt;/i&gt;&lt;.001; daughters: b=0.17, 95% CI 0.04-0.29, &lt;i&gt;P&lt;/i&gt;=.01). Decrease in social distancing by daughters was greater in the near-peer source group (b=-0.04, 95% CI -0.07 to 0.00, &lt;i&gt;P&lt;/i&gt;=.03) and lesser in the government agency group (b=0.05, 95% CI 0.02-0.09, &lt;i&gt;P&lt;/i&gt;=.003). The higher perceived credibility of the assigned information source increased social distancing (mothers: b=0.29, 95% CI 0.09-0.49, &lt;i&gt;P&lt;/i&gt;&lt;.01; daughters: b=0.31, 95% CI 0.11-0.51, &lt;i&gt;P&lt;/i&gt;&lt;.01) and vaccine intentions (mothers: b=4.18, 95% CI 1.83-6.53, &lt;i&gt;P&lt;/i&gt;&lt;.001; daughters: b=3.36, 95% CI 1.67-5.04, &lt;i&gt;P&lt;/i&gt;&lt;.001). Mothers' intentions to vaccinate self may have increased when they considered the near-peer source to be not credible (b=-0.50, 95% CI -0","PeriodicalId":73554,"journal":{"name":"JMIR infodemiology","volume":null,"pages":null},"PeriodicalIF":3.5,"publicationDate":"2022-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9400429/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"33446731","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Investigation of COVID-19 Misinformation in Arabic on Twitter: Content Analysis.","authors":"Ahmed Al-Rawi,&nbsp;Abdelrahman Fakida,&nbsp;Kelly Grounds","doi":"10.2196/37007","DOIUrl":"https://doi.org/10.2196/37007","url":null,"abstract":"<p><strong>Background: </strong>The COVID-19 pandemic has been occurring concurrently with an infodemic of misinformation about the virus. Spreading primarily on social media, there has been a significant academic effort to understand the English side of this infodemic. However, much less attention has been paid to the Arabic side.</p><p><strong>Objective: </strong>There is an urgent need to examine the scale of Arabic COVID-19 disinformation. This study empirically examines how Arabic speakers use specific hashtags on Twitter to express antivaccine and antipandemic views to uncover trends in their social media usage. By exploring this topic, we aim to fill a gap in the literature that can help understand conspiracies in Arabic around COVID-19.</p><p><strong>Methods: </strong>This study used content analysis to understand how 13 popular Arabic hashtags were used in antivaccine communities. We used Twitter Academic API v2 to search for the hashtags from the beginning of August 1, 2006, until October 10, 2021. After downloading a large data set from Twitter, we identified major categories or topics in the sample data set using emergent coding. Emergent coding was chosen because of its ability to inductively identify the themes that repeatedly emerged from the data set. Then, after revising the coding scheme, we coded the rest of the tweets and examined the results. In the second attempt and with a modified codebook, an acceptable intercoder agreement was reached (Krippendorff α≥.774).</p><p><strong>Results: </strong>In total, we found 476,048 tweets, mostly posted in 2021. First, the topic of infringing on civil liberties (n=483, 41.1%) covers ways that governments have allegedly infringed on civil liberties during the pandemic and unfair restrictions that have been imposed on unvaccinated individuals. Users here focus on topics concerning their civil liberties and freedoms, claiming that governments violated such rights following the pandemic. Notably, users denounce government efforts to force them to take any of the COVID-19 vaccines for different reasons. This was followed by vaccine-related conspiracies (n=476, 40.5%), including a Deep State dictating pandemic policies, mistrusting vaccine efficacy, and discussing unproven treatments. Although users tweeted about a range of different conspiracy theories, mistrusting the vaccine's efficacy, false or exaggerated claims about vaccine risks and vaccine-related diseases, and governments and pharmaceutical companies profiting from vaccines and intentionally risking the general public health appeared the most. Finally, calls for action (n=149, 12.6%) encourage individuals to participate in civil demonstrations. These calls range from protesting to encouraging other users to take action about the vaccine mandate. For each of these categories, we also attempted to trace the logic behind the different categories by exploring different types of conspiracy theories for each category.</p><p><strong>Conclus","PeriodicalId":73554,"journal":{"name":"JMIR infodemiology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9327499/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40576466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Infodemic Management Using Digital Information and Knowledge Cocreation to Address COVID-19 Vaccine Hesitancy: Case Study From Ghana.","authors":"Anna-Leena Lohiniva, Anastasiya Nurzhynska, Al-Hassan Hudi, Bridget Anim, Da Costa Aboagye","doi":"10.2196/37134","DOIUrl":"10.2196/37134","url":null,"abstract":"<p><strong>Background: </strong>Infodemic management is an integral part of pandemic management. Ghana Health Services (GHS) together with the UNICEF (United Nations International Children's Emergency Fund) Country Office have developed a systematic process that effectively identifies, analyzes, and responds to COVID-19 and vaccine-related misinformation in Ghana.</p><p><strong>Objective: </strong>This paper describes an infodemic management system workflow based on digital data collection, qualitative methodology, and human-centered systems to support the COVID-19 vaccine rollout in Ghana with examples of system implementation.</p><p><strong>Methods: </strong>The infodemic management system was developed by the Health Promotion Division of the GHS and the UNICEF Country Office. It uses Talkwalker, a social listening software platform, to collect misinformation on the web. The methodology relies on qualitative data analysis and interpretation as well as knowledge cocreation to verify the findings.</p><p><strong>Results: </strong>A multi-sectoral National Misinformation Task Force was established to implement and oversee the misinformation management system. Two members of the task force were responsible for carrying out the analysis. They used Talkwalker to find posts that include the keywords related to COVID-19 vaccine-related discussions. They then assessed the significance of the posts on the basis of the engagement rate and potential reach of the posts, negative sentiments, and contextual factors. The process continues by identifying misinformation within the posts, rating the risk of identified misinformation posts, and developing proposed responses to address them. The results of the analysis are shared weekly with the Misinformation Task Force for their review and verification to ensure that the risk assessment and responses are feasible, practical, and acceptable in the context of Ghana.</p><p><strong>Conclusions: </strong>The paper describes an infodemic management system workflow in Ghana based on qualitative data synthesis that can be used to manage real-time infodemic responses.</p>","PeriodicalId":73554,"journal":{"name":"JMIR infodemiology","volume":null,"pages":null},"PeriodicalIF":3.5,"publicationDate":"2022-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9281514/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40629818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Codeveloping and Evaluating a Campaign to Reduce Dementia Misconceptions on Twitter: Machine Learning Study.","authors":"Sinan Erturk,&nbsp;Georgie Hudson,&nbsp;Sonja M Jansli,&nbsp;Daniel Morris,&nbsp;Clarissa M Odoi,&nbsp;Emma Wilson,&nbsp;Angela Clayton-Turner,&nbsp;Vanessa Bray,&nbsp;Gill Yourston,&nbsp;Andrew Cornwall,&nbsp;Nicholas Cummins,&nbsp;Til Wykes,&nbsp;Sagar Jilka","doi":"10.2196/36871","DOIUrl":"https://doi.org/10.2196/36871","url":null,"abstract":"<p><strong>Background: </strong>Dementia misconceptions on Twitter can have detrimental or harmful effects. Machine learning (ML) models codeveloped with carers provide a method to identify these and help in evaluating awareness campaigns.</p><p><strong>Objective: </strong>This study aimed to develop an ML model to distinguish between misconceptions and neutral tweets and to develop, deploy, and evaluate an awareness campaign to tackle dementia misconceptions.</p><p><strong>Methods: </strong>Taking 1414 tweets rated by carers from our previous work, we built 4 ML models. Using a 5-fold cross-validation, we evaluated them and performed a further blind validation with carers for the best 2 ML models; from this blind validation, we selected the best model overall. We codeveloped an awareness campaign and collected pre-post campaign tweets (N=4880), classifying them with our model as misconceptions or not. We analyzed dementia tweets from the United Kingdom across the campaign period (N=7124) to investigate how current events influenced misconception prevalence during this time.</p><p><strong>Results: </strong>A random forest model best identified misconceptions with an accuracy of 82% from blind validation and found that 37% of the UK tweets (N=7124) about dementia across the campaign period were misconceptions. From this, we could track how the prevalence of misconceptions changed in response to top news stories in the United Kingdom. Misconceptions significantly rose around political topics and were highest (22/28, 79% of the dementia tweets) when there was controversy over the UK government allowing to continue hunting during the COVID-19 pandemic. After our campaign, there was no significant change in the prevalence of misconceptions.</p><p><strong>Conclusions: </strong>Through codevelopment with carers, we developed an accurate ML model to predict misconceptions in dementia tweets. Our awareness campaign was ineffective, but similar campaigns could be enhanced through ML to respond to current events that affect misconceptions in real time.</p>","PeriodicalId":73554,"journal":{"name":"JMIR infodemiology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9987190/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9421213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Social Listening to Enhance Access to Appropriate Pandemic Information Among Culturally Diverse Populations: Case Study From Finland.","authors":"Anna-Leena Lohiniva,&nbsp;Katja Sibenberg,&nbsp;Sara Austero,&nbsp;Natalia Skogberg","doi":"10.2196/38343","DOIUrl":"https://doi.org/10.2196/38343","url":null,"abstract":"<p><strong>Background: </strong>Social listening, the process of monitoring and analyzing conversations to inform communication activities, is an essential component of infodemic management. It helps inform context-specific communication strategies that are culturally acceptable and appropriate for various subpopulations. Social listening is based on the notion that target audiences themselves can best define their own information needs and messages.</p><p><strong>Objective: </strong>This study aimed to describe the development of systematic social listening training for crisis communication and community outreach during the COVID-19 pandemic through a series of web-based workshops and to report the experiences of the workshop participants implementing the projects.</p><p><strong>Methods: </strong>A multidisciplinary team of experts developed a series of web-based training sessions for individuals responsible for community outreach or communication among linguistically diverse populations. The participants had no previous training in systematic data collection or monitoring. This training aimed to provide participants with sufficient knowledge and skills to develop a social listening system based on their specific needs and available resources. The workshop design took into consideration the pandemic context and focused on qualitative data collection. Information on the experiences of the participants in the training was gathered based on participant feedback and their assignments and through in-depth interviews with each team.</p><p><strong>Results: </strong>A series of 6 web-based workshops was conducted between May and September 2021. The workshops followed a systematic approach to social listening and included listening to web-based and offline sources; rapid qualitative analysis and synthesis; and developing communication recommendations, messages, and products. Follow-up meetings were organized between the workshops during which participants could share their achievements and challenges. Approximately 67% (4/6) of the participating teams established social listening systems by the end of the training. The teams tailored the knowledge provided during the training to their specific needs. As a result, the social systems developed by the teams had slightly different structures, target audiences, and aims. All resulting social listening systems followed the taught key principles of systematic social listening to collect and analyze data and used these new insights for further development of communication strategies.</p><p><strong>Conclusions: </strong>This paper describes an infodemic management system and workflow based on qualitative inquiry and adapted to local priorities and resources. The implementation of these projects resulted in content development for targeted risk communication, addressing linguistically diverse populations. These systems can be adapted for future epidemics and pandemics.</p>","PeriodicalId":73554,"journal":{"name":"JMIR infodemiology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10014086/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9421219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}