{
  "id": "2009.14738",
  "version": "v1",
  "published": "2020-09-30T15:24:51.000Z",
  "updated": "2020-09-30T15:24:51.000Z",
  "title": "ResGCN: Attention-based Deep Residual Modeling for Anomaly Detection on Attributed Networks",
  "authors": [
    "Yulong Pei",
    "Tianjin Huang",
    "Werner van Ipenburg",
    "Mykola Pechenizkiy"
  ],
  "categories": [
    "cs.LG",
    "stat.ML"
  ],
  "abstract": "Effectively detecting anomalous nodes in attributed networks is crucial for the success of many real-world applications such as fraud and intrusion detection. Existing approaches have difficulties with three major issues: sparsity and nonlinearity capturing, residual modeling, and network smoothing. We propose Residual Graph Convolutional Network (ResGCN), an attention-based deep residual modeling approach that can tackle these issues: modeling the attributed networks with GCN allows to capture the sparsity and nonlinearity; utilizing a deep neural network allows to directly learn residual from the input, and a residual-based attention mechanism reduces the adverse effect from anomalous nodes and prevents over-smoothing. Extensive experiments on several real-world attributed networks demonstrate the effectiveness of ResGCN in detecting anomalies.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2020-09-30T15:24:51.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "attributed networks",
      "anomaly detection",
      "residual graph convolutional network",
      "attention-based deep residual modeling approach",
      "anomalous nodes"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}