A Novel Approximation to Dynamic Time Warping allows Anytime Clustering of Massive Time Series Datasets.
Qiang ZhuGustavo E. A. P. A. BatistaThanawin RakthanmanonEamonn J. KeoghPublished in: SDM (2012)
Keyphrases
- dynamic time warping
- distance measure
- shape retrieval
- lower bounding
- similarity measure
- quasi periodic
- data mining tasks
- clustering algorithm
- subsequence matching
- euclidean distance
- similarity search
- closed form
- self organizing maps
- elastic matching
- nearest neighbor classifier
- sequence matching
- edit distance
- word spotting
- pairwise
- object recognition
- data mining
- triangular inequality
- signature verification
- similarity function
- data points
- data analysis
- database systems