arXiv：cs.AI· Manar Alsaid·· 20 小时前
研究评估 AI 模型在社交噪声下检测虚假信息的表现，RoBERTa 准确率达 97.17%
Assessing the Performance of AI Models in Misinformation Detection Using Predefined Social Noise Constructs
AI 导读
一项研究用预设社交噪声构念评估六种文本模型和一个 GNN 在虚假信息检测中的表现，RoBERTa 准确率达 97.17%，Cohen's kappa、MCC 均为 0.9660，AUC 为 0.9977。
正文
Abstract:Social media platforms generate vast volumes of user-generated content that often obscure meaningful discourse, a phenomenon conceptualized as social noise. While existing misinformation detection methods primarily focus on factual verification, they largely overlook the role of social noise in amplifying misinformation. This study assesses the performance of six text-based models and a graph neural network (GNN) in mitigating the effects of misinformation using predefined social noise constructs. The dataset consists of 500,000 tweets collected from the X platform using the Ukraine hashtag during the Russia-Ukraine conflict from March 2024 to March 2025. Advanced machine learning models achieve strong performance, with accuracies ranging from 79.94% to 97.17%. RoBERTa demonstrates the highest performance, with an accuracy of 97.17%, Cohen's kappa of 0.9660, MCC of 0.9660, and AUC of 0.9977. The GNN achieves an accuracy of 64.52% and an AUC of 0.899, demonstrating performance above random-chance classification by exploiting the relational structure between tweets. All text-based models achieve kappa values greater than 0.75, indicating substantial overall agreement. Explainable AI analysis using LIME reveals that 89.6% of predictive weight is attributed to contextual features rather than isolated keywords. The findings highlight that integrating social noise into AI pipelines enhances misinformation detection and supports more effective mitigation strategies in large-scale social media environments.
| Comments: | 22 pages, 12 tables, 2 figures and one Appendix |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.17469 [cs.LG] |
| (or arXiv:2610.17469v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.17469 arXiv-issued DOI via DataCite (pending registration) |
|
| Journal reference: | ACM Transactions on Asian and Low-Resource Language Information Processing, 2026 |
Submission history
From: Manar Alsaid [view email]
[v1]
Mon, 12 Oct 2026 22:31:47 UTC (1,554 KB)
来源：arXiv：cs.AI · arxiv.org