A New Method for Estimating the Number of Distinct Values over Data Streams.
Longjiang GuoYingshu LiMeirui RenZhongzhao ZhangPublished in: SNPD (2009)
Keyphrases
- data streams
- pairwise
- significant improvement
- optimization algorithm
- data sets
- computationally efficient
- computational complexity
- preprocessing
- computational cost
- dynamic programming
- em algorithm
- classification accuracy
- clustering method
- synthetic data
- high precision
- sliding window
- matching algorithm
- streaming data
- model selection
- high accuracy
- probabilistic model
- experimental evaluation
- wireless sensor networks
- cost function
- feature vectors
- bayesian networks
- similarity measure