2003 · 30 citations · 0 references
Natural SciencesAstronomyPhysicsData MiningObservational AstronomyTime Series AnalysisAnomaly Detection In Time SeriesEvent DetectionGamma-ray BurstsGamma-ray Data AnalysisReal-time AnalyticsTime Series AnalyticsBurst DetectionBurst Detection AlgorithmsGamma-ray Burst AnalysisSliding Window AlgorithmsSliding Window AnalysisStream-based Event DetectionStreaming Anomaly Detection
Burst detection identifies abnormal aggregates in data streams using sliding windows, and in many applications it is necessary to monitor multiple window sizes simultaneously to flag windows with aggregates that differ significantly from other periods. The authors propose a general data structure that enables detection of interesting aggregates over elastic windows in near‑linear time. The data structure is applied to large‑scale astrophysical data for detecting Gamma Ray Bursts and to real‑time NYSE Trade and Quote data for identifying periods of high trading volume and volatility. The algorithm outperforms direct computation by several orders of magnitude in both astrophysical and financial data applications.
Burst detection is the activity of finding abnormal aggregates in data streams. Such aggregates are based on sliding windows over data streams. In some applications, we want to monitor many sliding window sizes simultaneously and to report those windows with aggregates significantly different from other periods. We will present a general data structure for detecting interesting aggregates over such elastic windows in near linear time. We present applications of the algorithm for detecting Gamma Ray Bursts in large-scale astrophysical data. Detection of periods with high volumes of trading activities and high stock price volatility is also demonstrated using real time Trade and Quote (TAQ) data from the New York Stock Exchange (NYSE). Our algorithm beats the direct computation approach by several orders of magnitude.