A Multi-Branch Feature Fusion Approach for Health Misinformation Detection and Propagation
本文提出一种多分支融合框架，结合语义、修辞线索及心理因素，以检测和分析在线社交网络中健康错误信息的传播，并引入认知传播评分辅助风险评估。
0 citationsRead paper
Selected work
Representative Papers
LDAC-Net: A Learnable Multi-Lag Differencing Attention-Convolution Network for Drift-Robust Recognition with Low-Cost MOX Gas Sensors
针对低成本MOX气体传感器信号受漂移等问题，提出LDAC-Net网络，通过学习多滞后差分和注意力卷积方法提高识别准确性。
0 citationsRead paper
Adaptive Memory and Reflection Multi-Agent System for Medical Question Answering
本文针对现有医疗问答系统缺乏适应性、持久记忆和结构化决策的问题，提出了一种基于多代理的自适应记忆与反思系统，通过专门的记忆和反馈机制提高复杂案例处理能力。
0 citationsRead paper
Ready for What? Rethinking AI and Robotics Preparedness for Adoption and Policy
研究通过分析982名参与者对17个AI和机器人挑战的评估，揭示了复杂性与准备度之间的关系，强调了在政策制定中应考虑具体挑战的差异。
0 citationsRead paper
Integrating Persuasion Theory into the Epidemiological Modelling of Health Misinformation Spread on Social Media
This study addresses the lack of individual psychological dynamics in modeling health misinformation propagation on social media by proposing the ELM-SIRMMM framework. This approach integrates psychological signals, including emotion and engagement from the Elaboration Likelihood Model (ELM), into a six-compartment epidemiological model to enable behavior-driven dynamic modulation of transmission rates. Multi-dataset validation demonstrates that the framework significantly enhances prediction accuracy and dynamic realism. Specifically, on the FibVID dataset, it reduces RMSE by 5.5% and corrects peak timing by 10 days. Furthermore, on MC-Fake, it accurately reproduces flash-rumor patterns with a 97% recovery rate, confirming the critical role of psychological mechanisms in achieving precise misinformation propagation modeling.
0 citationsRead paper
Recent publications
Latest Papers
A Multi-Branch Feature Fusion Approach for Health Misinformation Detection and Propagation
本文提出一种多分支融合框架，结合语义、修辞线索及心理因素，以检测和分析在线社交网络中健康错误信息的传播，并引入认知传播评分辅助风险评估。
0 citationsRead paper
LDAC-Net: A Learnable Multi-Lag Differencing Attention-Convolution Network for Drift-Robust Recognition with Low-Cost MOX Gas Sensors
针对低成本MOX气体传感器信号受漂移等问题，提出LDAC-Net网络，通过学习多滞后差分和注意力卷积方法提高识别准确性。
0 citationsRead paper
Adaptive Memory and Reflection Multi-Agent System for Medical Question Answering
本文针对现有医疗问答系统缺乏适应性、持久记忆和结构化决策的问题，提出了一种基于多代理的自适应记忆与反思系统，通过专门的记忆和反馈机制提高复杂案例处理能力。
0 citationsRead paper
Ready for What? Rethinking AI and Robotics Preparedness for Adoption and Policy
研究通过分析982名参与者对17个AI和机器人挑战的评估，揭示了复杂性与准备度之间的关系，强调了在政策制定中应考虑具体挑战的差异。
0 citationsRead paper
Integrating Persuasion Theory into the Epidemiological Modelling of Health Misinformation Spread on Social Media
This study addresses the lack of individual psychological dynamics in modeling health misinformation propagation on social media by proposing the ELM-SIRMMM framework. This approach integrates psychological signals, including emotion and engagement from the Elaboration Likelihood Model (ELM), into a six-compartment epidemiological model to enable behavior-driven dynamic modulation of transmission rates. Multi-dataset validation demonstrates that the framework significantly enhances prediction accuracy and dynamic realism. Specifically, on the FibVID dataset, it reduces RMSE by 5.5% and corrects peak timing by 10 days. Furthermore, on MC-Fake, it accurately reproduces flash-rumor patterns with a 97% recovery rate, confirming the critical role of psychological mechanisms in achieving precise misinformation propagation modeling.
0 citationsRead paper