The field of social media analysis and large language models is rapidly evolving, with a focus on developing innovative methods for analyzing and understanding online content. Recent studies have explored the use of large language models for tasks such as sentiment analysis, topic modeling, and bias detection, with applications in areas like election interference, hate speech detection, and climate change discourse. Notably, researchers are working to address the challenges of data access and bias in large language models, as well as developing new methodologies for evaluating and mitigating these issues. The development of datasets and frameworks for analyzing social media content, such as EDTok and UKElectionNarratives, is also underway. Some particularly noteworthy papers in this area include 'The Great Data Standoff: Researchers vs. Platforms Under the Digital Services Act', which provides a comprehensive analysis of the challenges of data access under the DSA, and 'Large Language Models are overconfident and amplify human bias', which highlights the need for further research into the limitations and biases of large language models. Additionally, 'Unraveling Media Perspectives: A Comprehensive Methodology' introduces a novel methodology for scalable, minimally biased analysis of media bias in political news.
Advances in Social Media Analysis and Large Language Models
Sources
The Great Data Standoff: Researchers vs. Platforms Under the Digital Services Act
Unraveling Media Perspectives: A Comprehensive Methodology Combining Large Language Models, Topic Modeling, Sentiment Analysis, and Ontology Learning to Analyse Media Bias
Amplifying Your Social Media Presence: Personalized Influential Content Generation with LLMs
UK Finfluencers: Exploring Content, Reach, and Responsibility
Automated Sentiment Classification and Topic Discovery in Large-Scale Social Media Streams
Large Language Models are overconfident and amplify human bias
EDTok: A Dataset for Eating Disorder Content on TikTok
Personalisation or Prejudice? Addressing Geographic Bias in Hate Speech Detection using Debias Tuning in Large Language Models
Towards High-Fidelity Synthetic Multi-platform Social Media Datasets via Large Language Models
Developing A Framework to Support Human Evaluation of Bias in Generated Free Response Text
Say It Another Way: A Framework for User-Grounded Paraphrasing
Doing Audits Right? The Role of Sampling and Legal Content Analysis in Systemic Risk Assessments and Independent Audits in the Digital Services Act
The Influence of Text Variation on User Engagement in Cross-Platform Content Sharing
Mapping the Climate Change Landscape on TikTok
Event-aware analysis of cross-city visitor flows using large language models and social media data
Large Language Models are often politically extreme, usually ideologically inconsistent, and persuasive even in informational contexts
To Judge or not to Judge: Using LLM Judgements for Advertiser Keyphrase Relevance at eBay
Large Means Left: Political Bias in Large Language Models Increases with Their Number of Parameters
Frame In, Frame Out: Do LLMs Generate More Biased News Headlines than Humans?
UKElectionNarratives: A Dataset of Misleading Narratives Surrounding Recent UK General Elections