A Unified Time Series Event Detection Framework
Detect events in time series
Anomaly detector using autoencoder
Anomaly detector based on machine learning classification
Anomaly detector using DTW
Anomaly detector using kmeans
Anomaly detector using ARIMA.
Anomaly detector using EMD
Anomaly detector using FBIAD
Anomaly detector using FFT
Anomaly detector using GARCH
Anomaly detector using histogram
Anomaly detector based on machine learning regression.
Anomaly and change point detector using RED
Anomaly detector using REMD
Anomaly detector using Wavelet
Harbinger Ensemble
Evaluation of event detection
Evaluation of event detection
Plot event detection on a time series
Harbinger
Harbinger Utils
At most one change (AMOC) method
Binary segmentation (BinSeg) method
Change Finder using ARIMA
Change Finder using ETS
Change Finder using LR
Chow test method
Change Finder using GARCH
Generalized Fluctuation Test (GFT)
Pruned exact linear time (PELT) method
Anomaly and change point detector using RED
Seminal change point
Discord discovery using Matrix Profile
Discord discovery using SAX
Motif discovery using Matrix Profile
Motif discovery using SAX
Motif discovery using xsax
Multivariate anomaly detector using PCA
Moving average smoothing
SAX
XSAX
By analyzing time series, it is possible to observe significant changes in the behavior of observations that frequently characterize events. Events present themselves as anomalies, change points, or motifs. In the literature, there are several methods for detecting events. However, searching for a suitable time series method is a complex task, especially considering that the nature of events is often unknown. This work presents Harbinger, a framework for integrating and analyzing event detection methods. Harbinger contains several state-of-the-art methods described in Salles et al. (2020) <doi:10.5753/sbbd.2020.13626>.