A Novel Method for Fast and Accurate Similarity Measure in Time Series Field.
Miaomiao ZhangDechang PiPublished in: ICDM Workshops (2017)
Keyphrases
- similarity measure
- experimental evaluation
- high accuracy
- detection method
- neural network
- high precision
- preprocessing
- significant improvement
- cost function
- classification accuracy
- completely automatic
- highly accurate
- segmentation method
- computationally efficient
- computational cost
- high quality
- support vector machine
- mutual information
- input data
- support vector machine svm
- detection algorithm
- clustering method
- dynamic programming
- prior knowledge
- classification method
- feature extraction
- moving average
- similarity computation
- feature selection