Structure, Topics, and Diffusion Effects of Bluesky Starter Packs

Profile-Free Behavioral Characterization of Bot-like Activity in a Political Reply Ecosystem on X: A Case Study

Coordinated and bot-like activity on social media is usually studied with supervised detectors that need rich account data such as profiles, timelines, and follower networks, which is increasingly hard to obtain. We ask what can be established about a single account’s reply ecosystem from its publicly visible posts and replies alone, with no profiles, timelines, or follower data. In a case study of the reply ecosystem of an official political party account (23,953 replies by 1985 accounts, December 2025 to January 2026), we compute profile-free behavioral features covering text duplication, character-level entropy, timing regularity, reply latency, and post coverage, complemented by a co-commenting network analysis, and group active accounts with unsupervised density-based clustering. The clustering, combined with two transparent labeling rules, separates three behavioral tiers: templated amplifiers defined by text reuse, persistent responders with human-like text but extreme volume and coverage, and an organic remainder. The two non-organic tiers comprise 5.4% of accounts, yet produce 53.5% of all comments, a composition that is stable under resampling and threshold sensitivity analysis, with a failure mode that is only conservative, since over-strict settings leave a tier unassigned rather than reshaping it. The platform’s own spam flags, never used as input, rise steadily from organic accounts to templated amplifiers, consistent with the behavioral grouping.

Shifts in economic discourse in online learning communities: evidence from the stack exchange

Influence Analysis of High-Follower Social Media Influencers

As social media continues to reshape consumer behavior, gauging the true reach of digital influencers has emerged as a pressing concern for scholars and industry professionals alike. Drawing upon data harvested from TikTok and Instagram in September 2022, this investigation probes the often-assumed link between audience size and content resonance. The TikTok corpus encompasses 1,000 profiles, tracking metrics from follower tallies to interaction patterns—views, likes, comments, and shares. Complementing this, the Instagram sample comprises 200 accounts, each profiled through proprietary influence scores and two-month engagement trajectories. Statistical scrutiny via correlation and regression modeling yields a nuanced picture: follower numbers correlate only moderately with engagement indicators (r = 0.36-0.46), and surprisingly, account for a mere 12.5% of variation in like counts. The platform divide proves equally striking—TikTok creators command engagement rates roughly threefold those on Instagram (7.2% versus 2.3%). For brands navigating the influencer landscape, these results underscore a pivotal insight: raw follower statistics tell only part of the story; engagement caliber and platform dynamics warrant equal, if not greater, attention.

Platform Power in News Aggregation

News aggregators are among the primary gatekeepers of online news, shaping both consumption and production through the algorithms that rank stories and sources. We provide the first large-scale estimates of this algorithmic power using the example of Yandex News, Russia's largest aggregator. Combining archived front pages with 12.3 million published articles and outlet-level traffic data, we estimate that occupying the entire top-5 news block for a day raises an outlet's traffic by roughly 380,000 visitors. After a 2016 law made aggregators liable for cited content, Yandex halved its references to independent outlets, yet consumers did not penalize the manipulation: its referrals converted readers at the same rate, and its market share did not fall. Removed outlets shifted toward longer articles and stopped optimizing headlines for the algorithm, showing that aggregator algorithms shape not only news consumption but also production.

Mapping the Reddit Bot Ecosystem: Taxonomy and Evolution

Automated agents increasingly participate in online communities, yet their population structure and roles remain poorly understood. Using a dataset of 3,389 identified bots and their full activity histories, we construct a taxonomy of bot"species"on the news aggregation and social media platform Reddit based on temporal, community, linguistic, and semantic features. Clustering analysis reveals 18 distinct bot types spanning content-specialized, behavior-driven, and infrastructural roles such as moderation and utility support. In addition, temporal analysis shows that bot numbers and activity expanded rapidly before peaking around the COVID-19 period, then started declining even before Reddit's 2023 API policy changes. However, the overall diversity of bot species has remained remarkably stable. These findings suggest that online bot populations form evolving digital ecosystems.

AI-Driven Recommender Systems, Filter Bubbles, and Romanian Facebook Users: A Qualitative Analysis

Filter bubbles describe the phenomenon in which algorithm‑driven personalization of online content narrows users’ exposure, insulating them from broader or opposing viewpoints. Recommender systems, used widely across most digital platforms, are the foundational technology underlying the creation of filter bubbles, as they are designed to predict preferences and suggest relevant items, being especially prone to creating such insulated environments. This paper investigates how Romanian Facebook users experience AI‑driven recommender systems, focusing on perceived filter bubbles, echo chambers, and their emotional and relational consequences. Based on 12 in depth interviews with current and former users, the study explores three questions: how users describe personalization and bubble effects in their feeds, how they experience the emotional and psychological impact of AI‑curated content, and what changes they envision to improve their social media experience. Respondents report highly personalized, repetitive feeds that bring forward political, news, and group content while partially ghosting everyday posts from friends. They specifically link bubble formation to both algorithmic curation and their own practices of unfollowing, muting, and engaging with content. Respondents describe complex emotional states, ranging from amusement and connection to anxiety, anger, envy, guilt, and feelings of addiction. Former users who reduced or quit Facebook report improved well‑being, greater focus, and more deliberate information consumption from alternative sources. In all interviews, users express a strong desire for greater control and transparency, including adjustable parameters for their news feed, better filtering of unwanted content, and stronger moderation of negative speech and disinformation. The study contributes to the understanding of how AI‑driven recommender systems impact Romanian users’ information environment, emotions, and social connections, highlighting both active coping strategies but also the limits of individual solutions.