arXiv subjects

Aleksandar Tomašević

Publications and source records attributed to Aleksandar Tomašević.

7 recordsLinked to original sources

arXiv2026

A Latent Oscillator Measurement Model to Simulate Emotional-Expression Score Dynamics in Video

Facial-expression classifiers convert video into multivariate time series of scores with measurement error from classifiers, videos, and recording conditions. Empirical score series cannot establish whether the score channels reflect a smaller set of latent expressive processes or whether an analysis would recover those processes. We introduce the Latent Oscillator Measurement Model (LOMM), a data-generating model that separates latent dynamics, time-varying activity, and a factor-analytic observation model. The latent processes are damped, undamped, or amplifying linear oscillators. LOMM generates bounded scores or continuous indicators. Study 1 used four-fold cross-fitting with scores from 100 MAFW videos to calibrate LOMM and evaluate generated series on held-out videos. Median plausibility and coverage were 0.970 and 0.920 for LOMM, versus 0.510 and 0.370 for a calibrated static logistic-normal generator. Study 2 tested whether Dynamic Exploratory Graph Analysis (DynEGA), static EGA, GraphicalVAR, and GIMME recovered a known dimensional structure from continuous indicators generated by LOMM. At 100 observations per clip, with failed or timed-out fits counted as incorrect, correct-dimension recovery was 0.939 for DynEGA, 0.884 for static EGA, 0.777 for GraphicalVAR, and 0.176 for GIMME. Replacing the common fixed initialization with independent stationary starts for stable dimensions and bounded independent starts for amplifying dimensions reduced recovery for DynEGA, static EGA, and GraphicalVAR. LOMM provides a controlled test of whether an analysis recovers aspecified latent structure before score patterns are interpreted psychologically.

physics.soc-ph↗

arXiv2026

Towards Operational Validation of LLM-Agent Social Simulations: A Replicated Study of a Reddit-like Technology Forum

Validation of LLM-agent social simulations remains underdeveloped, with most studies relying on subjective assessments or single runs. We address this gap by running 30 independent 30-day simulations of a technology forum modeled on Voat's v/technology, using stateless Dolphin Mistral 24B agents on the Y Social platform, and evaluating operational validity across five dimensions: activity patterns, network structure, toxicity, topical coverage, and stylistic convergence. Against 30 matched, non-overlapping 30-day Voat comparison windows, results show overlapping 99% confidence intervals for unique users, root posts, and daily active users, while comments, average thread length, and mean toxicity remain higher in simulation. Both simulated and empirical networks exhibit core-periphery structure, though simulated cores are larger and more diffuse and repeated interactions are less frequent. Topic alignment is near-complete, but toxicity is misallocated across content layers: simulated root posts are substantially more toxic than real submissions, while simulated comments are less toxic than Voat comments. These findings demonstrate that LLM agents in platform-faithful environments can reproduce familiar online regularities, while systematic divergences, particularly those linked to stateless agent design and content-layer calibration, point to concrete directions for future improvement.

cs.CY↗

arXiv2025

Reddit Deplatforming and Toxicity Dynamics on Generalist Voat Communities

Deplatforming, the permanent banning of entire communities, is a primary tool for content moderation on mainstream platforms. While prior research examines effects on banned communities or source platform health, the impact on alternative platforms that absorb displaced users remains understudied. We analyze four major Reddit ban waves (2015--2020) and their effects on generalist communities on Voat, asking how post-ban arrivals reshape community structure and through what mechanisms transformation occurs. Combining network analysis, toxicity detection, and dynamic reputation modeling, we identify two distinct regimes of migration impact: (1) Hostile Takeover (2015--2018), where post-ban arrival cohorts formed parallel social structures that bypassed existing community cores through sheer volume, and (2) Toxic Equilibrium (2018--2020), where the flattening of existing user hierarchy enabled newcomers to integrate into the now-dominant toxic community. Crucially, community transformation occurred through peripheral dynamics rather than hub capture: fewer than 5% of newcomers achieved central positions in most months, yet toxicity doubled. Migration structure also shaped outcomes: loosely organized communities dispersed into generalist spaces, while ideologically cohesive groups concentrated in dedicated enclaves. These findings suggest that receiving platforms face a narrow intervention window during the hostile takeover phase, after which toxic norms become self-sustaining.

cs.SI↗

arXiv2025

The Face of Populism: Examining Differences in Facial Emotional Expressions of Political Leaders Using Machine Learning

Populist rhetoric employed on online media is characterized as deeply impassioned and often imbued with strong emotions. The aim of this paper is to empirically investigate the differences in affective nonverbal communication of political leaders. We use a deep-learning approach to process a sample of 220 YouTube videos of political leaders from 15 different countries, analyze their facial expressions of emotion and then examine differences in average emotion scores representing the relative presence of 6 emotional states (anger, disgust, fear, happiness, sadness, and surprise) and a neutral expression for each frame of the YouTube video. Based on a sample of manually coded images, we find that this deep-learning approach has 53-60\% agreement with human labels. We observe statistically significant differences in the average score of negative emotions between groups of leaders with varying degrees of populist rhetoric.

cs.CY↗

arXiv2025

Multi-Platform Aggregated Dataset of Online Communities (MADOC)

The Multi-platform Aggregated Dataset of Online Communities (MADOC) is a comprehensive dataset that facilitates computational social science research by providing FAIR-compliant standardized access to cross-platform analysis of online social dynamics. MADOC aggregates and standardizes data from Bluesky, Koo, Reddit, and Voat (2012-2024), containing 18.9 million posts, 236 million comments, and 23.1 million unique users. The dataset enables comparative studies of toxic behavior evolution across platforms through standardized interaction records and sentiment analysis. By providing UUID-anonymized user histories and temporal alignment of banned communities' activity patterns, MADOC supports research on content moderation impacts and platform migration trends. Distributed via Zenodo with persistent identifiers and Python/R toolkits, the dataset adheres to FAIR principles while addressing post-API-era research challenges through ethical aggregation of public social media archives.

cs.SI↗

arXiv2022

Sustainability of Stack Exchange Q\&A communities: the role of trust

Knowledge-sharing communities are a fundamental element of any knowledge-based society. Understanding how they emerge, function, and disappear is thus of crucial importance. Many social and economic factors influence sustainable knowledge-sharing communities. Here we explore the role of the structure of social interactions and social trust in the emergence of these communities. Using tools from complex network theory, we analyze the early evolution of social structure in four pairs of StackExchange communities, each corresponding to one active and one closed community on the same topic. We adapt the dynamical reputation model to quantify the evolution of social trust in these communities. Our analysis shows that active communities have higher local cohesiveness and develop stable and more strongly connected cores. The average reputation is higher in sustainable communities. In these communities, the trust between core members develops early and remains high over time. Our results imply that efforts to create a stable and trustworthy core may be crucial for building a sustainable knowledge-sharing community.

physics.soc-ph↗

arXiv2022

Measuring the influence of beliefs in belief networks

Influential beliefs are crucial for our understanding of how people reason about political issues and make political decisions. This research proposes a new method for measuring the influence of political beliefs within larger context of belief system networks, based on the advances in psychometric network methods and network influence research. Using the latest round of the European Social Survey data, we demonstrate this approach on a belief network expressing support for the regime in 29 European countries and capturing beliefs related to support for regime performance, principles, institutions, and political actors. Our results show that the average influence of beliefs can be related to the consistency and connectivity of the belief network and that the influence of specific beliefs (e.g. Satisfaction with Democracy) on a country level has a significant negative correlation with external indicators from the same domain (e.g. Liberal Democracy index), which suggests that highly influential beliefs are related to pressing political issues. These findings suggest that network-based belief influence metrics estimated from large-scale survey data can be used a new type of indicator in comparative political research, which opens new avenues for integrating psychometric network analysis methods into political science methodology.

cs.SI↗