Structural Temporal Graph Neural Networks for Anomaly Detection in Dynamic Graphs

@article{Cai2021StructuralTG,
  title={Structural Temporal Graph Neural Networks for Anomaly Detection in Dynamic Graphs},
  author={Lei Cai and Zhengzhang Chen and Chen Luo and Jiaping Gui and Jingchao Ni and Ding Li and Haifeng Chen},
  journal={Proceedings of the 30th ACM International Conference on Information \& Knowledge Management},
  year={2021}
}
  • Lei Cai, Zhengzhang Chen, Haifeng Chen
  • Published 15 May 2020
  • Computer Science
  • Proceedings of the 30th ACM International Conference on Information & Knowledge Management
Detecting anomalies in dynamic graphs is a vital task, with numerous practical applications in areas such as security, finance, and social media. Existing network embedding based methods have mostly focused on learning good node representations, whereas largely ignoring the subgraph structural changes related to the target nodes in a given time window. In this paper, we propose StrGNN, an end-to-end structural temporal Graph Neural Network model for detecting anomalous edges in dynamic graphs… 

Figures and Tables from this paper

Anomaly Detection in Dynamic Graphs via Transformer
TLDR
This paper presents a novel Transformer-based Anomaly Detection framework for DYnamic graphs (TADDY), which constructs a comprehensive node encoding strategy to better represent each node’s structural and temporal roles in an evolving graphs stream via a dynamic graph transformer model.
Graph similarity learning for change-point detection in dynamic networks
TLDR
This work designs a method to perform online network change-point detection that can adapt to the network domain and localise changes with no delay and requires a shorter data history to detect changes than most existing state-of-the-art baselines.
A Comprehensive Survey on Graph Anomaly Detection with Deep Learning
TLDR
A taxonomy that follows a task-driven strategy and categorizes existing work according to the anomalous graph objects that they can detect is provided, and 12 future research directions spanning unsolved and emerging problems introduced by graph data, anomaly detection, deep learning and real-world applications are highlighted.
Foundations and Modeling of Dynamic Networks Using Dynamic Graph Neural Networks: A Survey
TLDR
This work establishes a foundation of dynamic networks with consistent, detailed terminology and notation and presents a comprehensive survey of dynamic graph neural network models using the proposed terminology.
Design of Graph-Based Layered Learning-Driven Model for Anomaly Detection in Distributed Cloud IoT Network
TLDR
This work designs an anomaly detection technique that attempts to efficiently monitor the entire network infrastructure to combat the spreading nature of cyber-attacks and proposes a multi-agent system that uses the collaborative environment of smart agents to find anomalies.
Hierarchical Graph Convolutional Network for Data Evaluation of Dynamic Graphs
TLDR
This work proposes a novel hierarchical graph convolutional network for the data evaluation of dynamic graphs, following the anomaly detection paradigm, and provides a more powerful tool for processing dynamic graphs.
FadMan: Federated Anomaly Detection across Multiple Attributed Networks
TLDR
The proposed algorithm FadMan is a vertical federated learning framework for public node aligned with many private nodes of different features, and is validated on two tasks — correlated anomalies detection on multiple attributed networks and anomaly detection on an attributeless network — using real-world datasets.
LUNAR: Unifying Local Outlier Detection Methods via Graph Neural Networks
TLDR
This paper proposes LUNAR, a novel, graph neural network-based anomaly detection method that performs significantly better than existing local outlier methods, as well as state-of-the-art deep baselines, and is much more robust to different settings of the local neighbourhood size.
Self-Supervised and Interpretable Anomaly Detection using Network Transformers
TLDR
The Network Transformer (NeT) is introduced, a DNN model for anomaly detection that incorporates the graph structure of the communication network in order to improve interpretability and provides a data-driven hierarchical approach to analyze the behavior of a cyber network.
AutoGML: Fast Automatic Model Selection for Graph Machine Learning
TLDR
This work develops the first meta-learning approach for automatic graph machine learning, called AUTOGML, which capitalizes on the prior performances of a large body of existing methods on benchmark graph datasets, and carries over this prior experience to automatically select an effective model to use for the new graph, without any model training or evaluations.
...
...

References

SHOWING 1-10 OF 67 REFERENCES
NetWalk: A Flexible Deep Embedding Approach for Anomaly Detection in Dynamic Networks
TLDR
This paper proposes a novel approach, NetWalk, for anomaly detection in dynamic networks by learning network representations which can be updated dynamically as the network evolves, and employs a clustering-based technique to incrementally and dynamically detect network anomalies.
Anomaly Detection using Graph Neural Networks
TLDR
The ability of Deep Learning over topological characteristics of a social network to detect anomalies in email network and twitter network is utilized and a model, Graph Neural Network, is presented, which is applied on social connection graphs to detect anomaly detection.
Graph based anomaly detection and description: a survey
TLDR
This survey aims to provide a general, comprehensive, and structured overview of the state-of-the-art methods for anomaly detection in data represented as graphs, and gives a general framework for the algorithms categorized under various settings.
DynGEM: Deep Embedding Method for Dynamic Graphs
TLDR
This work presents an efficient algorithm DynGEM, based on recent advances in deep autoencoders for graph embeddings, that can handle growing dynamic graphs, and has better running time than using static embedding methods on each snapshot of a dynamic graph.
Behavior-based Community Detection: Application to Host Assessment In Enterprise Information Networks
TLDR
A novel community detection framework is proposed to identify behavior-based host communities in enterprise information networks, purely based on large-scale heterogeneous event data, and an efficient method for assessing host's anomaly level by leveraging the detected host communities is proposed.
Community-based anomaly detection in evolutionary networks
TLDR
This work develops a parameter-free and scalable algorithm using a proposed representative-based technique to detect all six possible types of community-based anomalies: grown, shrunken, merged, split, born, and vanished communities, and details the underlying theory required to guarantee the correctness of the algorithm.
Network anomaly detection based on Eigen equation compression
TLDR
It is demonstrated through the experimental results using two benchmark data sets and a simulation data set that anomalies of a whole network and nodes responsible for them can be detected by the proposed network anomaly detection method.
Ranking Causal Anomalies via Temporal and Dynamical Analysis on Vanishing Correlations
TLDR
This paper proposes a network diffusion based framework that can effectively model fault propagation over the entire invariant network, and can perform joint inference on both the structural, and the time-evolving broken invariance patterns and can locate high-confidence anomalies that are truly responsible for the vanishing correlations.
Attentional Heterogeneous Graph Neural Network: Application to Program Reidentification
TLDR
An attentional heterogeneous graph neural network model (DeepHGNN) is proposed to verify the program's identity based on its system behaviors and developed an effective attentionalheterogeneous graph embedding algorithm to solve the program reidentification problem.
...
...