Yongwei Wang

Yongwei Wang

Auto-generated profile, based on publicly available information

Yongwei Wang

Zhengzhou Information Science and Technology Institute,Zhengzhou,China

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 August 2022 in arXiv
Verified
Preprint
Open access
PDF file
YWYongwei Wang
Yongwei Wang
YLYong Liu
Yong Liu
ZSZhiqi Shen
Zhiqi Shen
Graph neural networks (GNN) based collaborative filtering (CF) have attracted increasing attention in e-commerce and social media platforms. However, there still lack efforts to evaluate the robustness of such CF systems in deployment. Fundamentally different from existing attacks, this work revisits the item promotion task and reformulates it from a targeted topological attack perspective for the first time. Specifically, we first develop a targeted attack formulation to maximally increase a target item's popularity. We then leverage gradient-based optimizations to find a solution. However, we observe the gradient estimates often appear noisy due to the discrete nature of a graph, which leads to a degradation of attack ability. To resolve noisy gradient effects, we then propose a masked attack objective that can remarkably enhance the topological attack ability. Furthermore, we design a computationally efficient approach to the proposed attack, thus making it feasible to evaluate large-large CF systems. Experiments on two real-world datasets show the effectiveness of our attack in analyzing the robustness of GNN-based CF more practically.

29 October 2021 in Mathematical Problems in Engineering
Verified
Journal Article
Open access
PDF file
JLJing Liu
Jing Liu
CCChaoWen Chang
ChaoWen Chang
YZYuchen Zhang
Yuchen Zhang
YWYongwei Wang
Yongwei Wang
To address the problems of fusion efficiency, detection rate (DR), and false detection rate (FDR) that are associated with existing information fusion methods, a multisource information fusion method featuring dynamic evidence combination based on layer clustering and improved evidence theory is proposed in this study. First, the original alerts are hierarchically clustered and conflicting evidence is eliminated. Then, dynamic evidence combination is applied to fuse the condensed alerts, thereby improving the efficiency and accuracy of the fusion. The experimental results show that the proposed method is superior to current fusion methods in terms of fusion efficiency, DR, and FDR.

3 October 2022
Verified
Conference Paper
Closed access
YWYongwei Wang
Yongwei Wang
HSHuifang Su
Huifang Su
YZYuchen Zhang
Yuchen Zhang
YWYifei Wang
Yifei Wang
SZShengnan Zhou
Shengnan Zhou
PLPengcheng Liu
Pengcheng Liu
Evidence theory is one of the main methods to realize situational awareness. In this paper, the importance of evidence is measured and modified by dissimilarity calculation, and the measured value of dissimilarity calculation is taken as the basis of dynamic combination of evidence. A situation assessment method based on dissimilarity calculation and dynamic combination of evidence is proposed. The method includes four stages: rule metrics, evidence revision, rule fusion and situation decision. Experiments show that this method can avoid the paradoxical problem in situation fusion using evidence theory, and is superior to existing typical methods in terms of efficiency and accuracy of situation assessment.