A novel graph anomaly detection method called JPGFN (Feature Transformation Enhanced Jacobi Polynomial Graph Filtering Network), which significantly outperforms mainstream approaches on multiple real-world datasets.
Abstract
In recent years, graph anomaly detection (GAD) based on frequency-domain filtering have achieved promising results. However, existing approaches still face three major challenges: First, they use static basic function to constructed graph filter which cannot effectively adapt to the frequency-domain distribution of graph data. Second, they fail to adequately consider the importance information of each attribute in the node feature vector, leading to the loss of fine-grained information. Third, they insufficiently utilize node labels for GAD. To address these issues, this paper proposes a novel graph anomaly detection method called JPGFN (Feature Transformation Enhanced Jacobi Polynomial Graph Filtering Network). First, a Feature Separation Transformation Network (FSTNN) is developed to better learn fine-grained node features by feature separation and applying nonlinear transformations to node features across different dimensions. Second, an adaptive Jacobi polynomial graph filtering module is constructed based on Jacobi polynomials to adaptively capture complex frequency-domain features of graph signals. Finally, a node label constraint module is developed to facilitate the use of node labels and enhance the performance of GAD. Experimental results on multiple real-world datasets demonstrate that the proposed method significantly outperforms mainstream approaches.
LTRGAD is proposed, a two-stage GAD framework that performs feature selection based on local feature-topological residuals (LTR) and effectively introduces topological information while preserving the original local anomalous patterns, enabling more accurate local anomaly detection.
Yazheng Zhao, Nannan Wu, Hao Yin et al.· 0 citations
A novel framework, Generate and Filter graph learning for Graph Anomaly Detection (GFGAD), which generates a diverse set of synthetic anomalies with enriched feature and structural information to balance the data distribution and significantly outperforms state-of-the-art baselines.
Mengyu Li, Yonghao Liu, Ximing Li et al.· IEEE Transactions on Pattern...· 0 citations
The results indicate that entropy-based subgraph embedding can improve local anomaly detection performance, although the model does not achieve the best value for every metric on every dataset.
Gen Li, Jason J. Jung· Discover Computing· 0 citations
: Anomaly detection is used to identify data points deviating from normal patterns and plays a significant role in fields such as fault diagnosis, biomedicine, and cybersecurity. However, anomaly detection tasks in practical applications often involve high-dimensional data which presents challenges such as severe feature redundancy, hidden anomaly patterns, and difficulty in capturing uncertainty. These issues make it difficult to effectively identify anomalies, thereby limiting the model’s discriminative power and stability. To address these challenges, we propose a high-dimensional anomaly detection model, MSFO-EIF, integrating multi-strategy feature optimization with rough set modeling. First, in the feature space optimization stage, we incorporate label information and employ a Maximum Relevance Minimum Redundancy (mRMR) pre-screening and an uncertainty clustering mechanism to filter and structurally organize the original features, thereby reducing redundancy and retaining key discriminative information. Second, in the feature selection phase, we construct a multi-strategy co-evolution mechanism to optimize feature subsets within a label-guided search space, thereby mitigating the risk of local optima. Finally, in the anomaly detection stage, multi-region partitioning and rough set concepts are introduced to characterize the sample distribution structure at a fine granularity, thereby enhancing the ability to identify boundary samples and weak anomalies. Experimental results show that the proposed model outperforms other anomaly detection models, including KNN, LOF, IBBA-EIF, and RRSM, on multiple high-dimensional datasets. Specifically, MSFO-EIF achieves the improvements of 1.68%, 1.33%, 1.69%, 2.67% and 1.87% in accuracy, precision, F1-score, AUC and AUC-PR, respectively, highlighting its superior detection performance and robustness.
Dongfang Wu, Jiaojiao Deng, Zhiwei Ye et al.· Computers, Materials & C...· 0 citations
Video Anomaly Detection (VAD) is a crucial computer vision task for
security monitoring and public safety. Unsupervised VAD is more suitable for
real-world scenarios with rare unknown anomalies, but existing LLM-based
methods suffer from limited temporal modeling, inconsistent video understand
ing and inaccurate fine-grained localization, leading to biased anomaly scoring.
To solve these problems, we propose a novel unsupervised VAD framework fus
ing graph attention propagation and multimodal semantic information: first, fuse
video semantic and motion features to construct a dynamic spatiotemporal graph,
and refine node features via graph attention propagation with orthogonal con
straints; then, split videos into semantically coherent event units by a statistical
boundary detection module; finally, guide MLLMs to generate event semantic
descriptions and initial anomaly scores through a hierarchical prompting strategy,
and refine the scores via video-text semantic alignment to obtain accurate frame
level scores. Evaluated on UCF-Crime and XD-Violence datasets with frame
level AUC, the proposed framework achieves state-of-the-art performance under
unsupervised and zero-shot settings, significantly outperforming existing LLM
based VAD methods and even several weakly supervised approaches, which fully
verifies its effectiveness and robustness.
Qinghao Kong· Poster Volume 0008 The 2026...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
