基于多角度对比学习的属性网络异常检测

doi:10.3778/j.issn.1002-8331.2307-0011

摘要/Abstract

摘要： 传统属性网络的异常检测多基于自监督对比学习，存在着异常节点类型较少、对比学习角度单一等不足。基于“节点-节点”对比和“节点-子图”对比，提出了一种多角度对比学习属性网络异常检测的ADMC（anomaly detection on attribute network by multi-angle contrastive learning）模型。主要创新工作有：在原有结构和属性异常的基础上，进一步细化为四种节点异常类型，并在属性网络中对其进行量化测度；利用“节点-节点”的节点级对比获得二重联接信息，利用“节点-子图”的跨节点级对比获得局部连接信息，并构建了二者互为补充的多角度对比学习模型。基于社交媒体及引文文献等数据集，与基于自监督对比学习（CoLA）模型相比较的实验结果表明，ADMC模型能够丰富异常数据的类型，提升属性网络异常检测的精确性。

关键词: 属性网络, 异常检测, 对比学习

Abstract: Traditional anomaly detection in attribute networks largely relies on self-supervised contrastive learning, which often falls short due to the limited types of anomalous nodes and the singularity in contrastive learning perspectives. This paper proposes an ADMC (anomaly detection on attribute network by multi-angle contrastive learning) model based on “node-node” contrast and “node-subgraph” contrast, aiming to address these deficiencies. The primary innovations include: refining the original structural and attribute anomalies into four distinct node anomaly types and quantifying them within the attribute network; employing “node-node” node-level contrasts to acquire biconnected information, and utilizing “node-subgraph” cross-node level contrasts to obtain local connectivity information, thereby constructing a multi-angle contrastive learning model where both perspectives complement each other. Through experiments on datasets from social media and citation literature, compared with the self-supervised contrastive learning (CoLA) model, the results demons-trate that the ADMC model can enrich the types of anomaly data and enhance the precision of anomaly detection in attribute networks.

Key words: attribute network, anomaly detection, contrastive learning

李保珍, 孔倩文, 苏雨薇. 基于多角度对比学习的属性网络异常检测[J]. 计算机工程与应用, 2024, 60(19): 167-177.

LI Baozhen, KONG Qianwen, SU Yuwei. Anomaly Detection on Attribute Network by Multi-Angle Contrastive Learning[J]. Computer Engineering and Applications, 2024, 60(19): 167-177.

参考文献

[1] LIU Y, LI Z, PAN S, et al. Anomaly detection on attributed networks via contrastive self-supervised learning[J]. IEEE Transactions on Neural Networks and Learning Systems, 2022, 33(6): 2378-2392.
[2] LIU N, HUANG X, HU X. Accelerated local anomaly detection via resolving attributed networks[C]//Proceedings of the 26th International Joint Conference on Artificial Intelligence. Melbourne, Australia: International Joint Conferences on Artificial Intelligence Organization, 2017: 2337-2343.
[3] AKOGLU L, TONG H, KOUTRA D. Graph based anomaly detection and description: a survey[J]. Data Mining and Knowledge Discovery, 2015, 29(3): 626-688.
[4] XU Z, HUANG X, ZHAO Y, et al. Contrastive attributed network anomaly detection with data augmentation[C]//Proceedings of the 26th Pacific-Asia Conference on Knowledge Discovery and Data Mining. Berlin: Springer, 2022: 444-457.
[5] ROOPAK, SHAH, EDUARD, et al. Signature verification using a “siamese” time delay neural network[J]. International Journal of Pattern Recognition and Artificial Intelligence, 1993, 7(4): 669-688.
[6] BREUNIG M, KRIEGEL H P, NG R T, et al. LOF: identifying density-based local outliers[C]//Proceedings of the 2000 ACM SIGMOD International Conference on Management of Data. New York: ACM, 2000: 93-104.
[7] PEROZZI B, AKOGLU L. Scalable anomaly ranking of attributed neighborhoods[C]//Proceedings of the 2016 SIAM International Conference on Data Mining. Philadelphia: SIAM, 2016: 207-215.
[8] ZHANG M, CHEN Y. Weisfeiler-Lehman neural machine for link prediction[C]//Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2017: 575-583.
[9] 罗世杰, 吕文韬, 李凡, 等. 融合拓扑和属性的动态网络链路预测方法[J]. 计算机工程与应用, 2023, 59(5): 122-130.
LUO S J, LYU W T, LI F, et al. Dynamic network link prediction method for fusion topology and attributes[J]. Computer Engineering and Applications, 2023, 59(5): 122-130.
[10] GAO J, LIANG F, FAN W, et al. On community outliers and their efficient detection in information networks[C]//Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2010: 813-822.
[11] 吴越, 王英, 王鑫, 等. 基于超图卷积的异质网络半监督节点分类[J]. 计算机学报, 2021, 44(11): 2248-2260.
WU Y, WANG Y, WANG X, et al. Motif-based hypergraph convolution network for semi-supervised node classification on heterogeneous graph[J]. Chinese Journal of Computers, 2021, 44(11): 2248-2260.
[12] ABU-EL-HAIJA S, KAPOOR A, PEROZZI B, et al. N-GCN: multi-scale graph convolution for semi-supervised node classification[C]// Proceedings of the 35th Uncertainty in Artificial Intelligence Conference: Vol 115. New York: PMLR, 2020: 841-851.
[13] SCARSELLI F, GORI M, TSOI A C, et al. The graph neural network model[J]. IEEE Transactions on Neural Networks, 2009, 20(1): 61-80.
[14] 李忠, 靳小龙, 王亚杰, 等. 属性网络中基于变分图自编码器的异常节点检测方法[J]. 模式识别与人工智能, 2022, 35(1): 17-25.
LI Z, JIN X L, WANG Y J, et al. Anomaly node detection method based on variational graph auto-encoders in attribute networks [J]. Pattern Recognition and Artificial Intelligence, 2022, 35(1): 17-25.
[15] LI J, DANI H, HU X, et al. Radar: residual analysis for anomaly detection in attributed networks[C]//Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence. Melbourne, Australia: International Joint Conferences on Artificial Intelligence Organization, 2017: 2152-2158.
[16] ZHU D, MA Y, LIU Y. Anomaly detection with deep graph autoencoders on attributed networks[C]//Proceedings of the 2020 IEEE Symposium on Computers and Communications (ISCC). Piscataway: IEEE, 2020: 1-6.
[17] FAN H, ZHANG F, LI Z. AnomalyDAE: dual autoencoder for anomaly detection on attributed networks[C]//Proceedings of the 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Piscataway: IEEE, 2020: 5685-5689.
[18] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook, USA: Curran Associates Inc, 2017: 6000-6010.
[19] DING K, LI J, BHANUSHALI R, et al. Deep anomaly detection on attributed networks[C]//Proceedings of the 2019 SIAM International Conference on Data Mining. Philadelphia: SIAM, 2019: 594-602.
[20] KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks[J]. arXiv:1609.02907, 2016.
[21] 张伊扬, 钱育蓉, 陶文彬, 等. 基于深度学习的属性图异常检测综述[J]. 计算机工程与应用, 2022, 58(19): 1-13.
ZHANG Y Y, QIAN Y R, TAO W B, et al. Survey of attribute graph anomaly detection based on deep learning[J]. Computer Engineering and Applications, 2022, 58(19): 1-13.
[22] CHEN T, KORNBLITH S, NOROUZI M, et al. A simple framework for contrastive learning of visual representations[C]//Proceedings of the International Conference on Machine Learning. New York: PMLR, 2020: 1597-1607.
[23] VELICKOVIC P, FEDUS W, HAMILTON W L, et al. Deep graph infomax[J]. arXiv:1809.10341, 2018.
[24] PENG Z, HUANG W, LUO M, et al. Graph representation learning via graphical mutual information maximization[C]//Proceedings of the Web Conference 2020. New York: ACM, 2020: 259-270.
[25] ZHU Y, XU Y, YU F, et al. Graph contrastive learning with adaptive augmentation[C]//Proceedings of the Web Conference 2021. New York: ACM, 2021: 2069-2080.
[26] CHEN F, WANG Y C, WANG B, et al. Graph representation learning: a survey[J]. APSIPA Transactions on Signal and Information Processing, 2020, 9: e15.
[27] TANG J, QU M, WANG M, et al. LINE: large-scale information network embedding[C]//Proceedings of the 24th International Conference on World Wide Web, 2015: 1067-1077.
[28] TONG H, FALOUTSOS C, PAN J Y. Fast random walk with restart and its applications[C]//Proceedings of the Sixth International Conference on Data Mining (ICDM’06). Piscataway: IEEE, 2006: 613-622.
[29] TANG L, LIU H. Relational learning via latent social dimensions[C]//Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2009: 817-826.
[30] SEN P, NAMATA G, BILGIC M, et al. Collective classification in network data[J]. AI Magazine, 2008, 29(3): 93.
[31] TANG J, ZHANG J, YAO L, et al. ArnetMiner: extraction and mining of academic social networks[C]//Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2008: 990-998.
[32] JIN M, LIU Y, ZHENG Y, et al. ANEMONE: graph anomaly detection with multi-scale contrastive learning[C]//Proceedings of the 30th ACM International Conference on Information and Knowledge Management. New York: ACM, 2021: 3122-3126.
[33] DEECKE L, RUFF L, VANDERMEULEN R A, et al. Deep anomaly detection by residual adaptation[J]. arXiv:2010.02310, 2020.
[34] PANG G, SHEN C, JIN H, et al. Deep weakly-supervised anomaly detection[J]. arXiv:1910.13601, 2019.
[35] PENG Z, LUO M, LI J, et al. ANOMALOUS: a joint modeling approach for anomaly detection on attributed networks[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence. Palo Alto, CA: AAAI Press, 2018: 3513-3519.
[36] KINGMA D P, BA J. Adam: a method for stochastic optimization[J]. arXiv:1412.6980, 2014.