Towards Discovering the Intrinsic Cardinality and Dimensionality of Time Series Using MDL.
Bing HuThanawin RakthanmanonYuan HaoScott EvansStefano LonardiEamonn J. KeoghPublished in: Algorithmic Probability and Friends (2011)
Keyphrases
- intrinsic dimension
- minimum description length
- model selection
- dimensionality reduction
- high dimensional
- geometric structure
- dynamic time warping
- intrinsic dimensionality
- mdl principle
- information criterion
- multivariate time series
- neural network
- abnormal patterns
- moving average
- lower dimensional
- social networks
- learning algorithm
- data sets
- small size
- cardinality constraints
- sample size
- machine learning
- quasi periodic
- multidimensional time series