A Spark-Based Approach for High-Efficiency Embedded Feature Selection.
Fan ZhouZhongyang HanJun ZhaoWei WangPublished in: DASC/PiCom/DataCom/CyberSciTech (2019)
Keyphrases
- high efficiency
- feature selection
- high accuracy
- real and synthetic datasets
- memory space
- mutual information
- text categorization
- arbitrary shape
- feature space
- information gain
- machine learning
- feature selection algorithms
- text classification
- model selection
- feature set
- discriminative features
- small sample
- data sets
- unsupervised learning
- microarray data
- classification models
- multi class
- knn
- support vector
- result quality
- informative features
- forward selection