Searcharxiv⌕ Search

arXiv subjects

Ulya Bayram

Publications and source records attributed to Ulya Bayram.

3 recordsLinked to original sources

Continuous modeling of manipulative participation for stance analysis: Online discourse on the 2024 Street Animal Law in Türkiye

Studies of online political discourse commonly infer public opinion through stance analysis while implicitly assuming that observed user participation is authentic. Meanwhile, studies focusing on detecting manipulative participation are typically modeling it as a binary phenomenon, overlooking the varying degrees to which users may engage in coordinated or inauthentic behavior. This study proposes a continuous modeling approach for manipulative participation and examines its integration with stance analysis. Using online discourse on the 2024 Street Animal Law in Türkiye as a case study, we collected a dataset comprising 7,320,574 tweets from 537,994 users. From this corpus, 301 users were selected through a high-information sampling strategy and manually annotated by domain experts using a 10-criterion rubric that yields user-level continuous manipulative participation scores. Evaluation under a unified user-level cross-validation protocol showed that the proposed framework accurately estimates both manipulative participation and user stance, with gradient boosting methods providing the best manipulative participation estimates and BERTurk achieving the highest stance classification performance. External validation on an independently annotated user set demonstrated good generalization performance. The findings support modeling manipulative participation as a continuous construct rather than a binary label, providing a more informative basis for computational analyses of the relationship between manipulative participation and stance in online political discourse.

cs.CL↗

Rethinking Suicidal Ideation Detection: A Trustworthy Annotation Framework and Cross-Lingual Model Evaluation

Suicidal ideation detection is critical for real-time suicide prevention, yet its progress faces two under-explored challenges: limited language coverage and unreliable annotation practices. Most available datasets are in English, but even among these, high-quality, human-annotated data remains scarce. As a result, many studies rely on available pre-labeled datasets without examining their annotation process or label reliability. The lack of datasets in other languages further limits the global realization of suicide prevention via artificial intelligence (AI). In this study, we address one of these gaps by constructing a novel Turkish suicidal ideation corpus derived from social media posts and introducing a resource-efficient annotation framework involving three human annotators and two large language models (LLMs). We then address the remaining gaps by performing a bidirectional evaluation of label reliability and model consistency across this dataset and three popular English suicidal ideation detection datasets, using transfer learning through eight pre-trained sentiment and emotion classifiers. These transformers help assess annotation consistency and benchmark model performance against manually labeled data. Our findings underscore the need for more rigorous, language-inclusive approaches to annotation and evaluation in mental health natural language processing (NLP) while demonstrating the questionable performance of popular models with zero-shot transfer learning. We advocate for transparency in model training and dataset construction in mental health NLP, prioritizing data and model reliability.

cs.CL↗

Structural inequalities exacerbate infection disparities: A computational approach

During the COVID-19 pandemic, we witnessed a disproportionate infection rate among marginalized and low-income groups. Despite empirical evidence suggesting that structural inequalities in society contribute to health disparities, there has been little attempt to offer a computational and theoretical explanation to establish its plausibility and quantitative impact. This paper combines an agent-based network model and a compartmental susceptible-infectious-removed model (SIR) to explain how socioeconomic inequalities affect the dynamics of disease spreading. Specifically, we focus on two aspects of structural inequalities: wealth inequality and social segregation. Our computational model demonstrates that under high income inequality, the infection gap widens between the low-income and high-income groups, and also the overall infected cases increase. We also observed that social segregation between different socioeconomic status (SES) groups intensifies the spreading and hence mortality rates. Furthermore, we explain the second peak of the infectious cases during a pandemic as a result of a false sense of safety and loosening the quarantine, among the higher SES individuals. These findings send a strong message to policymakers; confinement measures must be accompanied by substantial financial assistance to those from lower-income groups so that the people regardless of their socioeconomic status can afford to stay in. Without financial assistance, lower-income individuals will remain in circulation, which will prolong the duration and magnitude of the infection.

physics.soc-ph↗