Change Point Detection Using Multivariate Exponentially Weighted Moving Average (MEWMA) for Optimal Parameter in Online Activity Monitoring.
Naveed KhanSally I. McCleanShuai ZhangChris D. NugentPublished in: UCAmI (1) (2016)
Keyphrases
- weighted moving average
- change point detection
- control charts
- activity monitoring
- optimal parameters
- statistical process control
- change point
- non stationary
- sequential data
- outlier detection
- normalized maximum likelihood
- parameter values
- cumulative sum
- process control
- information bottleneck method
- real time
- parameter settings
- genetic algorithm
- support vector regression
- multiscale
- fixed length
- adaptive control
- parallel algorithm
- business intelligence
- hidden markov models
- spatio temporal
- decision making
- neural network
- databases