{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/ranking-causal-anomalies-via-temporal-and","title":"Ranking Causal Anomalies via Temporal and Dynamical Analysis on Vanishing Correlations","arxiv_id":null,"date":"2016-07-19","proceeding":"ACM SIGKDD international conference on Knowledge discovery and data mining 2016 7","authors":["Wei Cheng","Kai Zhang","Haifeng Chen","Guofei Jiang","Zhengzhang Chen","Wei Wang"],"abstract":"Modern world has witnessed a dramatic increase in our ability to collect, transmit and distribute real-time monitoring\r\nand surveillance data from large-scale information systems and cyber-physical systems. Detecting system anomalies\r\nthus attracts significant amount of interest in many fields such as security, fault management, and industrial optimization. Recently, invariant network has shown to be a powerful way in characterizing complex system behaviours.\r\nIn the invariant network, a node represents a system component and an edge indicates a stable, significant interaction\r\nbetween two components. Structures and evolutions of the\r\ninvariance network, in particular the vanishing correlations, can shed important light on locating causal anomalies\r\nand performing diagnosis. However, existing approaches to\r\ndetect causal anomalies with the invariant network often\r\nuse the percentage of vanishing correlations to rank possible casual components, which have several limitations: 1)\r\nfault propagation in the network is ignored; 2) the root casual anomalies may not always be the nodes with a highpercentage of vanishing correlations; 3) temporal patterns\r\nof vanishing correlations are not exploited for robust detection. To address these limitations, in this paper we propose\r\na network diffusion based framework to identify significant causal anomalies and rank them. Our approach can effectively model fault propagation over the entire invariant\r\nnetwork, and can perform joint inference on both the structural, and the time-evolving broken invariance patterns. As\r\na result, it can locate high-confidence anomalies that are truly responsible for the vanishing correlations, and can compensate for unstructured measurement noise in the system.\r\nExtensive experiments on synthetic datasets, bank information system datasets, and coal plant cyber-physical system\r\ndatasets demonstrate the effectiveness of our approach.","url_abs":"https://www.kdd.org/kdd2016/papers/files/rfp0445-chengAemb.pdf","url_pdf":"https://www.kdd.org/kdd2016/papers/files/rfp0445-chengAemb.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"ranking-causal-anomalies-via-temporal-and","repo_url":"https://github.com/KnowledgeDiscovery/CausalRanking","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null},{"paper_slug":"ranking-causal-anomalies-via-temporal-and","repo_url":"https://github.com/chengw07/CausalRanking","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"management","task_name":"Management"},{"task_slug":"root-cause-ranking","task_name":"Root Cause Ranking"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}