Articles tagged with Social Media
A Penn State-developed algorithm can spot illicit online pharmacies that sell substandard medications and potentially addictive drugs without a prescription. The system uses web analytics and predictive models to identify bad sites and alert consumers before they make a purchase.
A team of researchers from Queensland University of Technology has developed an algorithm that can identify misogynistic content on Twitter with a high degree of accuracy. The algorithm uses a deep learning model to analyze language patterns and context, allowing it to distinguish between abusive language and innocent expressions.
Tribal behavior on social media widened the gulf between Remain and Leave voters in the UK's Brexit debate, with frequent use of derogatory terms and language. The study suggests that social media platforms' business model can exacerbate this polarization, making it challenging to create a more even-handed discussion.
A recent study found that FoMO is linked to loneliness, low self-esteem, and low self-compassion, not social media use. Researchers suggest addressing negative self-perceptions by practicing self-compassion and focusing on real connections.
A new analysis found that cerebral venous thrombosis (CVT) incidence has increased three times higher than previously reported in the US. CVT is most common in younger women, but recent findings suggest increases in men and older women, with a notable disparity in Black people compared to other racial and ethnic groups.
A new study from UC San Diego School of Medicine found nearly 2,000 fraudulent postings likely tied to fake COVID-19 health products, financial scams, and other consumer risks. The researchers used Natural Language Processing and machine learning to detect suspicious posts on Twitter and Instagram.
Researchers from the University of Copenhagen and Aalborg University found that people's eyes linger less on fake news headlines, spending a bit more time reading factual ones. This discovery could aid in fact-checking and help sort through the oceans of false information.
Researchers found that social media information can predict various personality traits and attributes, such as extraversion and IQ, using Twitter data. The study's machine learning approach may provide a new way for mental health diagnostics and personalized nudges.
A new study from Northumbria University reveals that the COVID-19 pandemic is having a profound negative impact on nine out of ten people with eating disorders. The research found that disruptions to daily life, social isolation, and reduced access to healthcare services exacerbated symptoms.
A study found that belief in negative vaccination misperceptions has a strong impact on policy views, with 20% of Americans holding negative views about vaccination. Strong support for mandatory childhood vaccinations was observed, but opposition to religious and personal belief exemptions remained significant.
A new study suggests that following an alternate Mediterranean or Alternative Healthy Eating Index diet pattern may be linked to fewer non-motor symptoms such as constipation, daytime sleepiness and depression. The researchers found a strong relationship between healthy diet adherence and reduced risk of these preceding symptoms.
Researchers analyzed genetic studies and found no causal link between sleep patterns and Alzheimer's disease. However, they discovered a small association between certain sleep characteristics and increased or decreased risk of developing Alzheimer's, including being a 'morning person' and having lower insomnia risk.
Researchers at the University of Texas at Austin created a new tool to help Facebook users discern real from fake news. The study found that combining two simple interventions, including a stop sign icon and a strong statement, had a significant effect on helping people think critically. The researchers suggest that Facebook can improv...
A new analysis reveals that rural America is experiencing a worsening shortage of emergency physicians, with 92% practicing in urban areas. The rural workforce is also aging, with nearly all recent graduates opting for urban practices, leaving many rural communities without adequate emergency care.
A novel strategy for quickly identifying Twitter trolls uses linguistic features, including repeated words and word pairs, to distinguish between genuine and deceptive messages. The algorithm requires as few as 50 tweets for accurate identification, offering a potential tool for combating hybrid warfare while preserving freedom of speech.
A study found that personal content, such as family photos and selfies, receives almost 17% more likes and 31% more comments than policy-focused posts. The study also shows that well-known congress members have significantly more followers.
A new study found that few TikTok videos address COVID-19 transmission and prevention, instead focusing on anxiety and quarantine. The researchers believe this missed opportunity to engage young people with vital health information related to the pandemic.
A new study by Columbia University researchers found that daily social media use is not a strong or consistent risk factor for depressive symptoms among adolescents. The study analyzed survey data from over 74,000 students and found no association between social media use and increased depressive symptoms.
A new study using big data and survey results from the 2018-19 flu season found strong associations between regional social media messages and vaccination attitudes and behavior. However, when people discussed vaccination with others, negative effects of social media on vaccine behavior disappeared.
A study published in Media International Australia found that COVID-19 misinformation led to a surge in conspiracy theories, particularly around 5G technology. The research highlighted the role of social media and celebrities in spreading false claims, which can have significant real-world impacts, such as attacks on mobile phone towers.
A study found that taking vitamin D and calcium supplements twice daily can reduce the chances of getting vertigo again, especially for those with low vitamin D levels. The study showed a 24% reduction in annual recurrence rate among participants who took supplements.
Researchers identified nine clusters of journalists, including an 'elite/legacy' cluster dominated by The Washington Post and others, which showed high levels of insularity. In contrast, a long-form/enterprise cluster fostered exchange among journalists with diverse perspectives.
Research suggests that increasing levels of education may be protective for people with a gene for familial early onset Alzheimer's, delaying the development of amyloid plaques. Participants with less than 10 years of education had twice as many plaques as those with more than 16 years.
Researchers found that autistic individuals tend to be more direct and precise in their emails, while non-autistic people are more likely to include social niceties. This study highlights the importance of adapting our email styles to be more respectful of others' differences.
A pilot study found that Snapchat improves diagnostic performance among radiology residents, who performed better on the platform than with traditional classroom screens. The results suggest a promising novel training method to enhance recognition of life-threatening imaging diagnoses.
A study of 8,275 participants found that those with severe gum disease had twice the risk for mild cognitive impairment or dementia. People with minimal tooth loss and mild gum disease were no more likely to develop thinking problems than those with no dental issues.
A study published by the American Academy of Neurology found that drivers who have had a concussion take longer to regain complex reaction times, crucial for driving situations. The research suggests that traditional reaction time tests may not be the best measure of driving responsiveness and readiness.
Research reveals complexity of tree and urban zone responses to extreme heating events. Tree transpiration plays a significant role in exacerbating the urban heat island effect.
Research highlights parallels between sci-fi author Frederik Pohl's works and modern world's consumer patterns, including hyper-consumption and ecological disaster. Dr Mike Ryder's study explores how Pohl's predictions are relevant today, with social media and augmented reality games mirroring his futuristic worlds.
A new study from the University of Colorado Anschutz Medical Campus finds that peer mentorship groups are better suited for fostering success among underrepresented groups. These groups provide flexible support and can be established online, making them an ideal option during the COVID-19 pandemic.
Research finds that frequent social media use can negatively impact depressive symptoms among LGBTQ youth, but a social media-free summer camp shows promise in reducing symptoms. After attending the camp, participants experienced a significant reduction in depressive symptoms, particularly those with high pre-camp social media use.
A study published in Neurology found that people with normal thinking and memory scores at the start of the study were less likely to develop later problems, even with amyloid plaques. Those with the APOE ε2 gene variant were also more resistant to developing plaques.
A team of researchers developed a technique to identify malicious internet accounts and track online foreign misinformation campaigns in real time. They found that their model was effective in identifying posts and accounts that were part of a foreign influence campaign, including those by accounts that had never been used before.
Researchers developed a machine learning system that can detect social media posts involved in coordinated political influence campaigns based on their content features. The system uses word count, webpage links, and posting time as digital fingerprints to identify suspicious activity.
Researchers found that college-aged participants spent more time on Snapchat, demonstrated more problematic use, and reported higher preference for negative social potency rewards. They also attempted to quit Facebook more than Snapchat, contradicting expectations.
Researchers at Cranfield University are exploring ways to detect and prevent insider threats, including bribed employees using internal tools. Their latest research focuses on analyzing employee reports to improve organizational security knowledge.
A study of 3267 men found that watching porn is associated with greater erectile dysfunction and lower satisfaction with sex. The research suggests that the type of porn watched may contribute to impaired erectile function, with 23% of men under-35 experiencing some level of erectile dysfunction.
A Rutgers study surveyed 134 college students of color and found that 90% observed police brutality videos, leading to symptoms of post-traumatic stress disorder, anxiety and fear of future police encounters. Researchers emphasize the need for police reform to stop killings of Black men, women and children.
A study of global tweets reveals that fear over COVID-19 has given way to anger, driven by xenophobia and isolation fatigue. Joy, gratitude, hope, and happiness also emerge, suggesting a sense of pride and community spirit.
Consumer-created social media visuals are reflective of consumer brand perceptions, according to researchers who developed a new model called BrandImageNet. This model maps images to perceptual attributes of a brand, allowing firms to automatically monitor consumers' brand perceptions and evaluate their positioning strategies.
A new study warns of the power of social influence in shaping organ donation decisions, revealing the 'lone wolf' effect where individuals opting out trigger a cascade of similar behavior. The researchers found that people are more likely to opt out if they perceive others doing so.
A new study reveals that aerosol boxes designed to protect healthcare workers during COVID-19 may not be effective and could even increase exposure to airborne particles. The research found that the boxes increased airborne particle contamination by a factor of 5 times or more, posing a significant risk to frontline health workers.
A study by MIT scholars reveals that social media sharing interferes with assessing news accuracy, but participants who think critically or have scientific knowledge share fewer false headlines. Rating non-Covid-19 headlines before sharing Covid-19 news increases accuracy significantly.
A recent study analyzing 10 years of HPV-vaccine related Facebook posts reveals that anti-vaxxers have gained a stronger voice against the use of the human papillomavirus vaccine. Negative posts displayed a 45% increase in public engagement, with some showing negative views on vaccine safety and effectiveness.
A new USC study has created a hate speech classifier that is more context-sensitive, reducing errors that amplify racial bias. The algorithm considers context and specific features of hate speech to improve accuracy.
Researchers used computer simulations and geological records to estimate magnetic field changes. They found that changes in the direction of the Earth's magnetic field may occur up to 10 times faster than previously thought, with rapid shifts associated with local weakening of the field.
A study analyzed 664 news articles about suicide posted on a social network between 2018 and 2019. The findings suggest that adherence to suicide reporting guidelines may benefit individual health as well as reader engagement, with US-based publications being less likely to adhere to the guidelines.
Researchers found that priming people to consider accuracy makes them more adept at recognizing truth from falsehoods, even if they don't think about it directly. By subtly nudging the brain to consider accuracy, individuals can improve their choices about what to share on social media.
The study highlights social media's benefits in Radiology, including educational content dissemination, professional networking, and collaboration. However, it also warns of challenges, such as maintaining patient privacy and avoiding e-professionalism.
A Columbia University research team has received a Facebook Probability and Programming research award to develop static analysis methods for probabilistic programming. The team aims to enhance the usability and accuracy of probabilistic programming by creating tools that automate model checking and compiler verification.
The COREnect project aims to develop a high-level strategic roadmap for core technologies in future connectivity systems and components. The roadmap will promote innovation and business opportunities, including for small and medium-sized enterprises, and contribute to building European leadership in 5G and beyond.
A University of Kansas study found that Kansans tend to rely on multiple sources for reliable COVID-19 information, with a preference for local authorities and distrust of social media. Information from government briefings, search engines, and national newspapers were among the most trusted sources.
A new study by University of Vaasa researchers highlights the importance of strategic social media use in creating customer engagement value. The study identifies four distinct social media marketing strategies: social commerce, content, monitoring, and CRM, which can foster collaborative conversations and enhance customer relationships.
Researchers analyzed over 11,000 Twitter posts to understand patient attitudes towards statins, finding that people perceive themselves as protected from harm while taking preventative medications. This study may inform public health messages and improve communication between healthcare professionals and patients.
Researchers at Facebook have developed an end-to-end system that converts 2D photos into 3D images, requiring no advanced photographic skills or equipment. The new technology provides seamless real-time conversion, robust enough to work on almost any photo, and generates results in seconds.
A study found that middle-aged women are more likely to have changes in the brain related to Alzheimer's disease, even when controlling for age and other factors. The researchers suggest that hormonal changes during menopause may play a role in this increased risk.
A new analysis of Twitter posts by members of the US Congress reveals rapid politicization of the COVID-19 pandemic, with polarization increasing over time. The study found that Democrats sent significantly more tweets about COVID-19 than Republicans, and their tweets were more likely to discuss public health and safety.
A study analyzing congressional tweets reveals deep divisions in COVID-19 messaging along party lines, with Democrats discussing the crisis more frequently and using different word choices than Republicans. The analysis found a stark contrast between Democratic and Republican communication on Twitter during the pandemic.
Research identifies highly motivated, altruistic individuals as key players in online health communities, supporting the management of long-term conditions. These superusers face difficulties but demonstrate autonomy and a desire to help others, highlighting their importance in healthcare.
A new study examines how mainstream media covers protests, finding that negative coverage is often more prevalent in anti-status quo events. The research suggests that journalists should focus on the substance of a protest rather than its violence or spectacle.