Data-driven Feature Selection for Long Longitudinal Breadth and High Dimensional Dataset: Empirical Studies of Metabolic Syndrome Prediction.
Ji-Han LiuCheng-Tse WuTa-Wei ChuJyh-Shing Roger JangPublished in: ICMLC (2020)
Keyphrases
- empirical studies
- data driven
- high dimensional
- feature selection
- uci datasets
- dimensionality reduction
- high dimensionality
- empirical analysis
- high dimensional datasets
- feature set
- prediction accuracy
- feature space
- real world data sets
- low dimensional
- microarray data
- small sample
- dimension reduction
- high dimension
- text classification
- microarray datasets
- machine learning
- prediction error
- prediction model
- gene expression data
- experimental design
- feature extraction
- text categorization
- variable selection
- multi class
- mutual information
- support vector
- synthetic datasets
- fold cross validation
- high dimensional data
- selected features
- information gain
- data sets
- model selection
- semi supervised
- feature vectors
- neural network