arXiv subjects

Ashfaq Ali Shafin

Publications and source records attributed to Ashfaq Ali Shafin.

10 recordsLinked to original sources

arXiv2026

Not All Duplicates Are Coordination: Generic vs. Non-Generic Duplicate Campaigns in Information Operations

Duplicate content is widely used to study coordinated behavior in social media information operations (IOs), but not all repetition provides equally meaningful evidence of coordination. Generic, reusable, or low-information posts may create noisy account-account links when projected into coordination graphs. We study this problem using 187,000 English-language tweets from six Russian Twitter Information Operations datasets. We introduce a generic/non-generic distinction for duplicate campaigns, label tweets using an LLM-assisted protocol with independent human validation, and train supervised classifiers over sentence embeddings to scale the labels. We construct duplicate campaigns using lexical similarity and two embedding-based methods. Generic campaigns are rare under lexical matching but account for nearly 39% of campaigns detected by embedding-based methods. Restricting graphs to non-generic campaigns reduces graph size and the largest connected component while increasing density, suggesting a smaller but more focused coordination structure. These findings show that duplicate-based coordination analysis should consider both textual similarity and semantic specificity.

cs.SI↗

arXiv2026

Leakage-Safe and Scheduler-Aware Machine Learning for Grid Job Runtime Prediction

Accurate job runtime prediction can improve scheduling-aware resource management in grid and distributed computing environments, but prediction models must be evaluated under realistic deployment constraints. This paper revisits CPU burst time prediction on the GWA-T-4 AuverGrid workload trace and reformulates it as leakage-safe pre-execution job runtime prediction. We define the target as job-level runtime, use only submission-time attributes, exclude post-execution variables, and evaluate models under temporal and cold-start settings rather than relying only on random cross-validation. We compare standard regressors, chronological historical baselines, categorical encoding strategies, and CatBoost with native categorical handling. We further add temporal hyperparameter tuning, runtime predictability analysis, feature ablation, error analysis by job length, and a minimal scheduling simulation. After temporal-validation tuning, CatBoost achieves the strongest deployment-oriented result with R^2=0.239, MAE=27,019, RMSE=46,587, and LogMAE=2.646 on the held-out temporal test set. A single-server simulation over all 69,523 held-out temporal test jobs shows that prediction-informed SJF reduces average waiting time by 50.92% relative to FCFS. The results show that random-split evaluation overestimates performance, categorical-native boosting improves temporal generalization, and long-job underestimation remains a scheduler-relevant challenge.

cs.LG↗

arXiv2026

PatchRisk: Forecasting Future Vulnerability Exposure in Open-Source Dependency Networks

Open-source software ecosystems are web-scale dependency networks. A downstream package can become exposed to security risk not because its own source code changes, but because one of its transitive dependencies later receives a vulnerability advisory. Existing vulnerability-detection work often focuses on whether code is currently vulnerable or whether a known vulnerable dependency is already present. We study a different problem: future transitive vulnerability exposure. Given a package-version dependency graph observed at release time, the task is to predict whether any non-root dependency will receive a vulnerability advisory within a future horizon. This problem is important for Web intelligence and software supply-chain security because it supports proactive dependency triage before future exposure is visible. It is also easy to evaluate incorrectly: current vulnerable dependencies can leak the label, and different versions of the same root package can create package-level memorization across train and test splits. We therefore construct PatchRisk, a leakage-aware benchmark from Open Source Vulnerability advisories and deps.dev dependency graphs for npm and PyPI. The benchmark uses filtration-aware labels, package-disjoint evaluation, temporal testing, and nested 1K, 3K, 5K, and 10K sampling scales. The largest cleaned setting contains 9,007 root package-version graphs spanning 4,157 root packages. We evaluate three feature families: TimeOnly, GraphStruct, and HistoryGraph. On the 10K temporal-test benchmark, HistoryGraph improves AUPRC over the strongest TimeOnly baseline from 0.351 to 0.640 for 90-day forecasting and from 0.473 to 0.813 for 365-day forecasting. The improvement remains stable across smaller scales and package-group shuffle robustness tests.

cs.CR↗

arXiv2026

Synthetic Data in Education: Empirical Insights from Traditional Resampling and Deep Generative Models

Synthetic data generation offers promise for addressing data scarcity and privacy concerns in educational technology, yet practitioners lack empirical guidance for selecting between traditional resampling techniques and modern deep learning approaches. This study presents the first systematic benchmark comparing these paradigms using a 10,000-record student performance dataset. We evaluate three resampling methods (SMOTE, Bootstrap, Random Oversampling) against three deep learning models (Autoencoder, Variational Autoencoder, Copula-GAN) across multiple dimensions: distributional fidelity (Kolmogorov-Smirnov distance, Jensen-Shannon divergence), machine learning utility such as Train-on-Synthetic-Test-on-Real scores (TSTR), and privacy preservation (Distance to Closest Record). Our findings reveal a fundamental trade-off: resampling methods achieve near-perfect utility (TSTR: 0.997) but completely fail privacy protection (DCR ~ 0.00), while deep learning models provide strong privacy guarantees (DCR ~ 1.00) at significant utility cost. Variational Autoencoders emerge as the optimal compromise, maintaining 83.3% predictive performance while ensuring complete privacy protection. We also provide actionable recommendations: use traditional resampling for internal development where privacy is controlled, and VAEs for external data sharing where privacy is paramount. This work establishes a foundational benchmark and practical decision framework for synthetic data generation in learning analytics.

cs.LG↗

arXiv2025

Duplicating Deceit: Inauthentic Behavior Among Indian Misinformation Duplicators on X/Twitter

This paper investigates inauthentic duplication on social media, where multiple accounts share identical misinformation tweets. Leveraging a dataset of misinformation verified by AltNews, an Indian fact-checking organization, we analyze over 12 million posts from 5,493 accounts known to have duplicated such content. Contrary to common assumptions that bots are primarily responsible for spreading false information, fewer than 1\% of these accounts exhibit bot-like behavior. We present TweeXster, a framework for detecting and analyzing duplication campaigns, revealing clusters of accounts involved in repeated and sometimes revived dissemination of false or abusive content.

cs.SI↗

arXiv2025

The Language of Influence: Sentiment, Emotion, and Hate Speech in State Sponsored Influence Operations

State-sponsored influence operations (SIOs) have become a pervasive and complex challenge in the digital age, particularly on social media platforms where information spreads rapidly and with minimal oversight. These operations are strategically employed by nation-state actors to manipulate public opinion, exacerbate social divisions, and project geopolitical narratives, often through the dissemination of misleading or inflammatory content. Despite increasing awareness of their existence, the specific linguistic and emotional strategies employed by these campaigns remain underexplored. This study addresses this gap by conducting a comprehensive analysis of sentiment, emotional valence, and abusive language across 2 million tweets attributed to influence operations linked to China, Iran, and Russia, using Twitter's publicly released dataset of state-affiliated accounts. We identify distinct affective and rhetorical patterns that characterize each nation's digital propaganda. Russian campaigns predominantly deploy negative sentiment and toxic language to intensify polarization and destabilize discourse. In contrast, Iranian operations blend antagonistic and supportive tones to simultaneously incite conflict and foster ideological alignment. Chinese activities emphasize positive sentiment and emotionally neutral rhetoric to promote favorable narratives and subtly influence global perceptions. These findings reveal how state actors tailor their information warfare tactics to achieve specific geopolitical objectives through differentiated content strategies.

cs.SI↗

arXiv2025

Toxicity in State Sponsored Information Operations

State-sponsored information operations (IOs) increasingly influence global discourse on social media platforms, yet their emotional and rhetorical strategies remain inadequately characterized in scientific literature. This study presents the first comprehensive analysis of toxic language deployment within such campaigns, examining 56 million posts from over 42 thousand accounts linked to 18 distinct geopolitical entities on X/Twitter. Using Google's Perspective API, we systematically detect and quantify six categories of toxic content and analyze their distribution across national origins, linguistic structures, and engagement metrics, providing essential information regarding the underlying patterns of such operations. Our findings reveal that while toxic content constitutes only 1.53% of all posts, they are associated with disproportionately high engagement and appear to be strategically deployed in specific geopolitical contexts. Notably, toxic content originating from Russian influence operations receives significantly higher user engagement compared to influence operations from any other country in our dataset. Our code is available at https://github.com/shafin191/Toxic_IO.

cs.SI↗

arXiv2025

Victim-Centred Abuse Investigations and Defenses for Social Media Platforms

Online abuse, a persistent aspect of social platform interactions, impacts user well-being and exposes flaws in platform designs that include insufficient detection efforts and inadequate victim protection measures. Ensuring safety in platform interactions requires the integration of victim perspectives in the design of abuse detection and response systems. In this paper, we conduct surveys (n = 230) and semi-structured interviews (n = 15) with students at a minority-serving institution in the US, to explore their experiences with abuse on a variety of social platforms, their defense strategies, and their recommendations for social platforms to improve abuse responses. We build on study findings to propose design requirements for abuse defense systems and discuss the role of privacy, anonymity, and abuse attribution requirements in their implementation. We introduce ARI, a blueprint for a unified, transparent, and personalized abuse response system for social platforms that sustainably detects abuse by leveraging the expertise of platform users, incentivized with proceeds obtained from abusers.

cs.CR↗

arXiv2023

THOS: A Benchmark Dataset for Targeted Hate and Offensive Speech

Detecting harmful content on social media, such as Twitter, is made difficult by the fact that the seemingly simple yes/no classification conceals a significant amount of complexity. Unfortunately, while several datasets have been collected for training classifiers in hate and offensive speech, there is a scarcity of datasets labeled with a finer granularity of target classes and specific targets. In this paper, we introduce THOS, a dataset of 8.3k tweets manually labeled with fine-grained annotations about the target of the message. We demonstrate that this dataset makes it feasible to train classifiers, based on Large Language Models, to perform classification at this level of granularity.

cs.CL↗

arXiv2022

Strategies and Vulnerabilities of Participants in Venezuelan Influence Operations

Studies of online influence operations, coordinated efforts to disseminate and amplify disinformation, focus on forensic analysis of social networks or of publicly available datasets of trolls and bot accounts. However, little is known about the experiences and challenges of human participants in influence operations. We conducted semi-structured interviews with 19 influence operations participants that contribute to the online image of Venezuela, to understand their incentives, capabilities, and strategies to promote content while evading detection. To validate a subset of their answers, we performed a quantitative investigation using data collected over almost four months, from Twitter accounts they control. We found diverse participants that include pro-government and opposition supporters, operatives and grassroots campaigners, and sockpuppet account owners and real users. While pro-government and opposition participants have similar goals and promotion strategies, they differ in their motivation, organization, adversaries and detection avoidance strategies. We report the Patria framework, a government platform for operatives to log activities and receive benefits. We systematize participant strategies to promote political content, and to evade and recover from Twitter penalties. We identify vulnerability points associated with these strategies, and suggest more nuanced defenses against influence operations.

cs.SI↗