A5048633988- Misinformation and Its Impacts
- Social Media and Politics
- Hate Speech and Cyberbullying Detection
- Sentiment Analysis and Opinion Mining
- Advanced Text Analysis Techniques
- Stock Market Forecasting Methods
- Topic Modeling
- Opinion Dynamics and Social Influence
- Digital Marketing and Social Media
- Spam and Phishing Detection
- Complex Network Analysis Techniques
- Natural Language Processing Techniques
- Auditing, Earnings Management, Governance
- Media Influence and Health
- Freedom of Expression and Defamation
- Financial Markets and Investment Strategies
- Market Dynamics and Volatility
- Housing Market and Economics
- Big Data and Business Intelligence
- Internet Traffic Analysis and Secure E-voting
- Digital Mental Health Interventions
- Reinforcement Learning in Robotics
- Media Influence and Politics
- Monetary Policy and Economic Impact
- Technology Adoption and User Behaviour
Giessen School of Theology
2019-2025
Justus-Liebig-Universität Gießen
2015-2025
University of Freiburg
2015-2023
University of Oxford
2018-2019
Science Oxford
2018
Abstract Online media is important for society in informing and shaping opinions, hence raising the question of what drives online news consumption. Here we analyse causal effect negative emotional words on consumption using a large dataset viral stories. Specifically, conducted our analyses series randomized controlled trials ( N = 22,743). Our comprises ~105,000 different variations stories from Upworthy.com that generated ∼5.7 million clicks across more than 370 overall impressions....
Abstract The Russian invasion of Ukraine in February 2022 was accompanied by practices information warfare, yet existing evidence is largely anecdotal while large-scale empirical lacking. Here, we analyze the spread pro-Russian support on social media. For this, collected $N = 349{,}455$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mi>N</mml:mi> <mml:mo>=</mml:mo> <mml:mn>349</mml:mn> <mml:mo>,</mml:mo> <mml:mn>455</mml:mn> </mml:math> messages from Twitter with support....
Misinformation undermines the credibility of social media and poses significant threats to modern societies. As a countermeasure, Twitter has recently introduced "Birdwatch," community-driven approach address misinformation on Twitter. On Birdwatch, users can identify tweets they believe are misleading, write notes that provide context tweet rate quality other users' notes. In this work, we empirically analyze how interact with new feature. For purpose, collect all Birdwatch ratings between...
While false rumors pose a threat to the successful overcoming of COVID-19 pandemic, an understanding how diffuse in online social networks is – even for non-crisis situations still its infancy. Here we analyze large sample consisting rumor cascades from Twitter that have been fact-checked by third-party organizations. The data comprises N = 10,610 retweeted more than 24 million times. We investigate whether misinformation spreads viral truth and differences diffusion true vs. can be...
The spread of misinformation on social media is a pressing societal problem that platforms, policymakers, and researchers continue to grapple with. As countermeasure, recent works have proposed employ non-expert fact-checkers in the crowd fact-check content. While experimental studies suggest crowds might be able accurately assess veracity content, an understanding how fact-checked (mis-)information spreads missing. In this work, we empirically analyze misleading vs. not community posts...
Deploying links to professional fact-checking websites (so-called “snoping”) is a common misinformation intervention technique that can be used by social media users refute misleading claims made others. However, the real-world effect of snoping may limited as it suffers from low visibility and distrust towards fact-checkers. As remedy, X (formerly known Twitter) recently launched its community-based system “Community Notes” on which fact-checks are carried out actual directly shown...
Abstract Emotions are regarded as a dominant driver of human behavior, and yet their role in online rumor diffusion is largely unexplored. In this study, we empirically study the extent to which emotions explain rumors. We analyze large-scale sample 107,014 rumors from Twitter, well cascades. For each rumor, embedded were measured based on eight so-called basic Plutchik’s wheel (i.e., anticipation–surprise, anger–fear, trust–disgust, joy–sadness). then estimated using generalized linear...
Abstract False rumors (often termed “fake news”) on social media pose a significant threat to modern societies. However, potential reasons for the widespread diffusion of false have been underexplored. In this work, we analyze whether sentiment words, as well different emotional in content explain differences spread true vs. rumors. For purpose, collected $${\varvec{N}} =126{,}301$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mi>N</mml:mi> </mml:mrow>...
Understanding emerging threats from social media platforms.
Social media platforms disseminate extensive volumes of online content, including true and, in particular, false rumors. Previous literature has studied the diffusion offline rumors, yet more research is needed to understand In this paper, we examine role lifetime and crowd effects social sharing behavior for vs. Based on 126,301 Twitter cascades, find that characterized by explain differences spread as opposed All else equal, a longer associated with less activities, reduction larger than...
Online hate speech is responsible for violent attacks such as, e.g., the Pittsburgh synagogue shooting in 2018, thereby posing a significant threat to vulnerable groups and society general. However, little known about what makes on social media go viral. In this paper, we collect N = 25,219 cascades with 65,946 retweets from X (formerly as Twitter) classify them hateful vs. normal. Using generalized linear regression, then estimate differences spread of normal content based author variables....
The storming of the U.S. Capitol on January 6, 2021 has led to killing 5 people and is widely regarded as an attack democracy. was largely coordinated through social media networks such Twitter "Parler". Yet little known regarding how users interacted Parler during Capitol. In this work, we examine emotion dynamics with regard heterogeneity across time users. For this, segment user base into different groups (e.g., Trump supporters QAnon supporters). We use affective computing infer emotions...
Community-based fact-checking is a promising approach to fact-check social media content at scale. However, an understanding of whether users trust community fact-checks missing. Here, we presented
Social media ads have become a key communication channel in politics. However, the relationship between political from social and election outcomes is not fully understood. Here, we aim to estimate association online advertising during 2021 German federal election. For this, analyze large-scale dataset of 21,641 Facebook Instagram that received ~126 million impressions. Using regression analysis, show on has positive with candidate's outcome may even sway elections. All else equal, ~200,000...
Social media ads have become a key communication channel in politics. However, the relationship between political from social and election outcomes is not fully understood. Here, we aim to estimate association online advertising during 2021 German federal election. For this, analyze large-scale dataset of 21,641 Facebook Instagram that received ≈126 million impressions. Using regression analysis, show on has positive with candidate's outcome may even sway elections. All else equal, ≈200,000...
Sentiment analysis refers to the extraction of polarity source materials, such as financial news. However, measuring positive tone requires correct classification sentences that are negated, i.e. The negation scopes. For example, around 4.74% all in German ad hoc announcements contain negations. To predict corresponding scope, related literature commonly utilizes two approaches, namely, rule-based algorithms and machine learning. Nevertheless, a thorough comparison is missing, especially for...
Social media has become an indispensable channel for political communication. However, the discourse is increasingly characterized by hate speech, which affects not only reputation of individual politicians but also functioning society at large. In this work, we empirically analyze how amount speech in replies to posts from on Twitter depends personal characteristics, such as their party affiliation, gender, and ethnicity. For purpose, employ Twitter's Historical API collect every tweet...
Fake news on social media has large, negative implications for society. However, little is known about what linguistic cues make people fall fake and, hence, how to design effective countermeasures media. In this study, we seek understand which news. Linguistic (e.g., adverbs, personal pronouns, positive emotion words, words) are important characteristics of any text and also affect process real vs. Specifically, compare the role across both cognitive processing (related careful thinking)...
ABSTRACT This article provides a holistic study of how stock prices vary in their response to financial disclosures across different topics. Thereby, we specifically shed light into the extensive amount filings for which no priori categorization content exists. For this purpose, utilize an approach from data mining—namely, latent Dirichlet allocation (LDA)—as means topic modeling. technique facilitates our task automatically categorizing, ex ante , more than 70,000 regulatory 8‐K U.S....