An Abnormal Pattern Detection Scheme Based on GCN and DBSCAN in a Large-Scale Graph
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
In recent decades, anomaly detection has undoubtedly become one of the most important areas of research. This is because applications such as financial transactions, medical fraud, and anomaly detection can be used to solve a wide range of real-life problems. Data from these applications can be modeled using large graphs of many different nodes and edges. Because of the size and heterogeneity of the data contained in the graph, it is a very difficult task to detect abnormal patterns. In this paper, we proposed a method for detecting abnormal patterns in a large homogeneous graph. The proposed method consisted of two steps. In the first step, the graph was transformed into a vector using a semi-supervised graph neural network (GCN). The second step was based on DBSCAN, an unsupervised clustering method. Various performance evaluations were performed to show the superiority of the proposed method. Experimental results showed that the proposed method could detect abnormal nodes with high accuracy in homogeneous static graphs.
Abstract
In recent decades, anomaly detection has undoubtedly become one of the most important areas of research. This is because applications such as financial transactions, medical fraud, and anomaly detection can be used to solve a wide range of real-life problems. Data from these applications can be modeled using large graphs of many different nodes and edges. Because of the size and heterogeneity of the data contained in the graph, it is a very difficult task to detect abnormal patterns. In this paper, we proposed a method for detecting abnormal patterns in a large homogeneous graph. The proposed method consisted of two steps. In the first step, the graph was transformed into a vector using a semi-supervised graph neural network (GCN). The second step was based on DBSCAN, an unsupervised clustering method. Various performance evaluations were performed to show the superiority of the proposed method. Experimental results showed that the proposed method could detect abnormal nodes with high accuracy in homogeneous static graphs.
참고문헌 (22)
- G. Malewiczm, "Pregel: a system for largescale graph processing", Proceedings of the 2010 ACM SIGMOD International Conference on Management of data, pp.135~146 (2010)
- R. F. Erbacher, "Intrusion and misuse detection in large-scale systems", IEEE computer graphics and applications, 22(1), pp.38~47 (2002)[DOI]
- J. Wang, "Fdgars: Fraudster detection via graph convolutional networks in online app review system", Companion Proceedings of the 2019 World Wide Web Conference (2019)
- R. Yu, "A survey on social media anomaly detection", ACM SIGKDD Explorations Newsletter, 18(1), pp.1~14 (2016)[DOI]
- S. Velampalli, "Discovering suspicious patterns using a graph-based approach", The Thirty-Second International Flairs Conference (2019)
- L. Akoglu, "Anomaly detection in large graphs", School of Computer Scuence Carnegie Mellon University Pittsburgh (2009)
- X. Ma, "A comprehensive survey on graph anomaly detection with deep learning", IEEE Transactions on Knowledge and Data Engineering (2021)[DOI]
- P. Kavehzadeh, "Unsupervised Anomaly Detection on Node Attributed Networks: A Deep Learning Approach", 4th International Conference on Information Science and Systems, ICISS 2021. Association for Computing Machinery (ACM) (2021)
- T. N. Kipf, "Semi-supervised classification with graph convolutional networks", Conference paper at ICLR 2017 (2017)
- M. Ester, "A density-based algorithm for discovering clusters in large spatial databases with noise", Proceedings of the Second International Conference on Knowledge Discovery and Data Mining, pp.226~231 (1996)
- W. Yu, "Netwalk: A flexible deep embedding approach for anomaly detection in dynamic networks", Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining (2018)
- E. A. Manzoor, "Fast memory-efficient anomaly detection in streaming heterogeneous graphs", Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2016)
- S. Bhatia, "Sketch-Based Streaming Anomaly Detection in Dynamic Graphs", Computer science, 2, pp.1~12 (2022)[DOI]
- M. Xu, "Understanding graph embedding methods and their applications", SIAM Review, 63(4), pp.825~853 (2021)[DOI]
- J. Chen, "Fastgcn: fast learning with graph convolutional networks via importance sampling", Conference: International Conference on Learning Representations (2018)
- "science", ~ (2018)
- S. Wang, "Graph Neural Networks: Foundations, Frontiers, and Applications", pp.557~578 (2022), Springer
- D. K. Xu, "A comprehensive survey of clustering algorithms", Annals of Data Science, 2, pp.165~193 (2015)[DOI]
- A. Sreenivasulu, "Evaluation of Cluster Based Anomaly Detection", Master, pp.2019
- J. Sander, "Density-based clustering in spatial databases: The algorithm gdbscan and its applications", Data mining and knowledge discovery, 2(2), pp.169~194 (1998)[DOI]
- N. Rahmah, "Determination of optimal epsilon (eps) value on dbscan algorithm to clustering data on peatland hotspots in Sumatra", IOP conference series: earth and environmental science, 31, pp.012012~ (2016)