Summary of Cluster Aware Graph Anomaly Detection, by Lecheng Zheng et al.
Cluster Aware Graph Anomaly Detection
by Lecheng Zheng, John R. Birge, Haiyue Wu, Yifang Zhang, Jingrui He
First submitted to arxiv on: 15 Sep 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed CARE (Cluster Aware Multi-view Graph Anomaly Detection) method addresses the challenges of view heterogeneity and label scarcity in graph anomaly detection. It captures local and global node affinities by augmenting the adjacency matrix with pseudo-labels, without strong assumptions about graph structures. The similarity-guided loss is introduced to mitigate potential biases from pseudo-labels, which is a variant of contrastive learning loss connected to graph spectral clustering. Experimental results on several datasets demonstrate CARE’s effectiveness and efficiency, outperforming competitors by over 39% on the Amazon dataset and 18.7% on the YelpChi dataset. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary CARE is a new way to find anomalies in big data that have different types of information. This is useful for finding fraud or insider threats in things like e-commerce platforms and cybersecurity systems. The problem is that these datasets often have many different kinds of information, which makes it hard to detect anomalies. Traditional methods don’t work well with this kind of data. CARE uses a special way to look at the data and find patterns that are not normal. It’s very good at finding anomalies and doing it quickly. |
Keywords
» Artificial intelligence » Anomaly detection » Spectral clustering
