信息安全研究2026，Vol.12Issue(5)：420-427,8.DOI:10.12379/j.issn.2096-1057.2026.05.04
基于图神经网络和多特征融合的有害网站检测研究
Research on Harmful Website Detection Based on Graph Neural Network and Multi-feature Fusion
摘要
Abstract
To address the limitations of current harmful website detection methods in deep text semantic mining and multimodal feature co-perception,this study proposes a multi-feature fusion detection model based on graph attention networks(GAT)and ConvNeXt.The framework leverages GloVe word embeddings to construct semantic representations of website text,mapping it into a graph structure based on word co-occurrence relationships.The adaptive attention mechanism in GAT dynamically captures contextual dependencies between non-contiguous words,while ConvNeXt extracts both local details and global contextual features from website images.A cross-attention-based fusion module facilitates dynamic text-image feature alignment and interactive integration.Experimental results demonstrate that the proposed model achieves 99.10％accuracy in four-category website classification,significantly enhancing detection performance.This work offers valuable insights for identifying harmful online content and enhancing cybersecurity governance.关键词
有害网站检测/图神经网络/多特征融合/图注意力网络/ConvNeXt/交叉注意力机制Key words
harmful website detection/graph neural network/multi-feature fusion/GAT/ConvNeXt/cross-attention分类
信息技术与安全科学引用本文复制引用
瞿淼樟,师智斌,常赵宇,张薇..基于图神经网络和多特征融合的有害网站检测研究[J].信息安全研究,2026,12(5):420-427,8.基金项目
信息网络安全公安部重点实验室(公安部第三研究所)开放课题(C23600-06) （公安部第三研究所）
