MERLIN: Parameter-Free Discovery of Arbitrary Length Anomalies in Massive Time Series Archives.
Takaaki NakamuraMakoto ImamuraRyan MercerEamonn J. KeoghPublished in: ICDM (2020)
Keyphrases
- video sequences
- parameter free
- arbitrary length
- categorical data
- multivariate time series data
- multivariate time series
- anomaly detection
- moving objects
- outlier detection
- digital libraries
- motif discovery
- knowledge discovery
- metadata
- b tree
- detect anomalies
- multimedia
- rare events
- cultural heritage
- dynamic time warping
- non stationary
- fully automatic
- data analysis
- data structure
- stock market
- database
- data mining tasks
- query processing
- object recognition
- image segmentation
- clustering algorithm
- databases