A GAN-Based Framework for Robust DDoS Attack Detection
本文针对DDoS攻击检测问题，提出了一种结合生成对抗网络和高级机器学习模型的鲁棒框架，通过生成合成对抗流量来增强模型对未知攻击的防御能力。
0 citationsRead paper
Selected work
Representative Papers
Entropy-Centric Explainable AI for Remote Sensing Image Segmentation
This study addresses the lack of transparency in deep model decisions for semantic segmentation of remote sensing imagery, which hinders their trustworthy deployment in critical applications. To this end, the work proposes the first explainable artificial intelligence (XAI) method tailored to remote sensing segmentation, centered on information entropy, and introduces a dedicated region-wise relevance evaluation paradigm to quantitatively assess the alignment between explanation outcomes and predicted semantics. Experimental results demonstrate that the proposed approach significantly outperforms existing XAI techniques adapted for segmentation tasks in terms of explanation fidelity, offering a novel pathway toward interpretable and reliable intelligent interpretation of remote sensing data.
0 citationsRead paper
Automated Big Data Quality Assessment using Knowledge Graph Embeddings
This study addresses the challenge of achieving context-aware precision in big data quality assessment, which existing methods struggle to accomplish. The authors propose a novel approach based on knowledge graph embeddings that integrates multi-source contextual information by modeling datasets, quality dimensions, and validation rules as a knowledge graph. To enable dynamic evaluation, numeric-valued edges are introduced to weight quality metrics adaptively according to contextual relevance. Leveraging the AmpliGraph framework, the method predicts missing relationships within the graph to automatically generate tailored data quality assessment plans. Experimental validation on real-world radiation sensor data from the Lebanese Atomic Energy Commission demonstrates that the proposed technique effectively produces comprehensive and accurate data quality evaluations, confirming its feasibility and innovation.
0 citationsRead paper
Explanations Leak: Membership Inference with Differential Privacy and Active Learning Defense
This work reveals that counterfactual explanations (CFs), while enhancing transparency in Machine Learning as a Service (MLaaS), inadvertently expand the attack surface for membership inference attacks, thereby exacerbating privacy risks. Specifically, the study demonstrates for the first time that CFs obtained via API queries significantly strengthen shadow-model-based membership inference attacks. To mitigate this vulnerability, the authors propose a unified defense framework integrating differential privacy with active learning. This approach effectively curbs privacy leakage while preserving both model utility and explanation quality. Experimental results show that the proposed method achieves a favorable trade-off among privacy protection, model performance, and interpretability, offering a novel pathway toward secure and trustworthy MLaaS deployments.
0 citationsRead paper
The First Crypto President: Presidential Power and Cryptocurrency Markets During Trump's Second Term (2025-2029)
This study examines the deep interconstitutive mechanisms among presidential authority, familial business interests, and cryptocurrency markets during Donald Trump’s second term (2025–2029). Method: Employing a mixed-methods approach—quantitative event study coupled with institutional qualitative analysis—it identifies causal linkages between policy signals and market responses. Contribution/Results: The paper introduces the novel concept of “politically affiliated digital assets,” demonstrating how executive influence—exerted via family-linked token launches, policy rhetoric, and regulatory relaxation—drives capital flows and triggers global cascading liquidations. Empirically, the Trump-aligned crypto ecosystem peaked at over $11 billion in market capitalization; however, the October 2025 tariff policy announcement precipitated a single-day loss exceeding $1 trillion, underscoring systemic conflicts of interest and structural market fragility. The study advances a new analytical framework and risk-forecasting paradigm at the intersection of political economy and digital finance.
0 citationsRead paper
Recent publications
Latest Papers
A GAN-Based Framework for Robust DDoS Attack Detection
本文针对DDoS攻击检测问题，提出了一种结合生成对抗网络和高级机器学习模型的鲁棒框架，通过生成合成对抗流量来增强模型对未知攻击的防御能力。
0 citationsRead paper
Entropy-Centric Explainable AI for Remote Sensing Image Segmentation
This study addresses the lack of transparency in deep model decisions for semantic segmentation of remote sensing imagery, which hinders their trustworthy deployment in critical applications. To this end, the work proposes the first explainable artificial intelligence (XAI) method tailored to remote sensing segmentation, centered on information entropy, and introduces a dedicated region-wise relevance evaluation paradigm to quantitatively assess the alignment between explanation outcomes and predicted semantics. Experimental results demonstrate that the proposed approach significantly outperforms existing XAI techniques adapted for segmentation tasks in terms of explanation fidelity, offering a novel pathway toward interpretable and reliable intelligent interpretation of remote sensing data.
0 citationsRead paper
Automated Big Data Quality Assessment using Knowledge Graph Embeddings
This study addresses the challenge of achieving context-aware precision in big data quality assessment, which existing methods struggle to accomplish. The authors propose a novel approach based on knowledge graph embeddings that integrates multi-source contextual information by modeling datasets, quality dimensions, and validation rules as a knowledge graph. To enable dynamic evaluation, numeric-valued edges are introduced to weight quality metrics adaptively according to contextual relevance. Leveraging the AmpliGraph framework, the method predicts missing relationships within the graph to automatically generate tailored data quality assessment plans. Experimental validation on real-world radiation sensor data from the Lebanese Atomic Energy Commission demonstrates that the proposed technique effectively produces comprehensive and accurate data quality evaluations, confirming its feasibility and innovation.
0 citationsRead paper
Explanations Leak: Membership Inference with Differential Privacy and Active Learning Defense
This work reveals that counterfactual explanations (CFs), while enhancing transparency in Machine Learning as a Service (MLaaS), inadvertently expand the attack surface for membership inference attacks, thereby exacerbating privacy risks. Specifically, the study demonstrates for the first time that CFs obtained via API queries significantly strengthen shadow-model-based membership inference attacks. To mitigate this vulnerability, the authors propose a unified defense framework integrating differential privacy with active learning. This approach effectively curbs privacy leakage while preserving both model utility and explanation quality. Experimental results show that the proposed method achieves a favorable trade-off among privacy protection, model performance, and interpretability, offering a novel pathway toward secure and trustworthy MLaaS deployments.
0 citationsRead paper
The First Crypto President: Presidential Power and Cryptocurrency Markets During Trump's Second Term (2025-2029)
This study examines the deep interconstitutive mechanisms among presidential authority, familial business interests, and cryptocurrency markets during Donald Trump’s second term (2025–2029). Method: Employing a mixed-methods approach—quantitative event study coupled with institutional qualitative analysis—it identifies causal linkages between policy signals and market responses. Contribution/Results: The paper introduces the novel concept of “politically affiliated digital assets,” demonstrating how executive influence—exerted via family-linked token launches, policy rhetoric, and regulatory relaxation—drives capital flows and triggers global cascading liquidations. Empirically, the Trump-aligned crypto ecosystem peaked at over $11 billion in market capitalization; however, the October 2025 tariff policy announcement precipitated a single-day loss exceeding $1 trillion, underscoring systemic conflicts of interest and structural market fragility. The study advances a new analytical framework and risk-forecasting paradigm at the intersection of political economy and digital finance.
0 citationsRead paper