网络与信息安全学报2024，Vol.10Issue(5)：163-174,12.DOI:10.11959/j.issn.2096-109x.2024075
基于数据增强的多视图对比学习图异常检测
Data augmentation based multi-view contrastive learning graph anomaly detection
摘要
Abstract
Graph anomaly detection is valuable in preventing harmful events such as financial fraud and network in-trusion.Although contrast-based anomaly detection methods could effectively mine anomaly information based on the inconsistency of anomalous node instance pairs,avoiding the drawback of using self-coding architecture that led to the need for full graph training for the model.However,most existing contrast-based graph anomaly detec-tion methods focused only on node-subgraph contrast patterns,ignoring the fact that the sampled node-subgraph in-stance pairs contained only the local information of the target node,and at the same time did not take into account the importance of each subgraph to the target node,which led to the lack of global information about the node and the emergence of the problem that the contrast patterns were too generalized.In order to solve the problems men-tioned above and to improve the accuracy of graph anomaly detection,a graph anomaly detection by data augmen-tation and multi-view contrastive learning(DAMC-GAD)was proposed.Specifically,a graph data augmentation method for anomaly detection was proposed,in which the relative local structure of target nodes and their own attri-butes were used to correlate distant nodes in order to construct an augmented view that was rich in global informa-tion.Layer-by-layer sampling combined with node-subgraph contrast was introduced,and optimization strategies for the contrast model were developed based on the importance of the subgraph.A multi-view contrastive learning model with data augmentation was constructed through a complementary fusion strategy,and extensive experi-ments were conducted on synthetic anomaly and real anomaly datasets,which show that DAMC-GAD outperforms the current state-of-the-art baseline model on both types of datasets.关键词
数据增强/多视图对比学习/图神经网络/图异常检测Key words
data augmentation/multi-view contrastive learning/graph neural network/graph anomaly detection分类
信息技术与安全科学引用本文复制引用
李一凡,李家印,林兴澎,戴远飞,许力..基于数据增强的多视图对比学习图异常检测[J].网络与信息安全学报,2024,10(5):163-174,12.基金项目
国家自然科学基金(62471139,U1905211,62202221) （62471139,U1905211,62202221）
国家科技项目备案-中央引导地方科技发展专项(2023L3007) （2023L3007）
江苏省自然科学基金青年项目(BK20220331) （BK20220331）
福建省自然科学基金(2023J05128) （2023J05128）
福建省科技创新重点项目(2022G02003) The National Natural Science Foundation of China(62471139,U1905211,62202221),The National Science and Technology Project Record-Centralized Guided Local Science and Technology Development Special Project(2023L3007),The Natural Science Foundation of Jiangsu Province(BK20220331),The Natural Science Foundation of Fujian Province(2023J05128),Science and Technology Projects in Fujian Province(2022G02003) （2022G02003）
