Villa-Cox Ramon, Williams Evan M, Carley Kathleen M
ESPAE, Escuela Superior Politecnica del Litoral, Guayaquil, Guayas, Ecuador.
Software and Societal Systems, Carnegie Mellon University, Pittsburgh, Pennsylvania, United States of America.
PLoS One. 2025 Jun 26;20(6):e0324697. doi: 10.1371/journal.pone.0324697. eCollection 2025.
Stance detection is an important task with a wide range of high-impact social applications, including opinion polling and detecting propaganda, misinformation, and hate speech. In this work, we explore the performance and extrapolation power of political stance-detection models using an existing large-scale weakly-labeled Twitter dataset collected around the 2019 South American Protests. We construct transformer-based user and tweet encoders to embed users in a low-dimensional space using their posts and interactions. We then train heterogeneous graph attention networks to predict user stances and contrast their ability to extrapolate stance predictions to different country contexts and to future events. We find that leveraging users' ego-network in political stance detection improves in-country model performance for every country we examine. More notably, we find that leveraging a user's social context greatly enhances the ability of our stance detection models to extrapolate to new country contexts and future data.
立场检测是一项重要任务，具有广泛的高影响力社会应用，包括民意调查以及检测宣传、错误信息和仇恨言论。在这项工作中，我们使用围绕2019年南美抗议活动收集的现有大规模弱标签推特数据集，探索政治立场检测模型的性能和外推能力。我们构建基于Transformer的用户和推文编码器，以便利用用户的帖子和互动将其嵌入到低维空间中。然后，我们训练异构图注意力网络来预测用户立场，并对比它们将立场预测外推到不同国家背景和未来事件的能力。我们发现，在我们研究的每个国家，在政治立场检测中利用用户的自我网络可提高国内模型的性能。更值得注意的是，我们发现利用用户的社会背景极大地增强了我们的立场检测模型外推到新国家背景和未来数据的能力。
