Using causal discovery for feature selection in multivariate numerical time series.
Youqiang SunJiuyong LiJixue LiuChristopher ChowBing-Yu SunRujing WangPublished in: Mach. Learn. (2015)
Keyphrases
- causal discovery
- feature selection
- markov blanket
- multivariate time series
- causal models
- causal relationships
- discovery process
- bayesian networks
- causal structure
- conditional independence
- text categorization
- high dimensionality
- directed acyclic graph
- feature set
- classification accuracy
- dimension reduction
- text classification
- machine learning
- observational data
- structural equation models
- support vector
- structure learning
- feature subset
- causal relations
- sensitivity analysis
- naive bayes
- knowledge discovery
- real world