A new feature selection algorithm based on binomial hypothesis testing for spam filtering.
Jieming YangYuanning LiuZhen LiuXiaodong ZhuXiaoxu ZhangPublished in: Knowl. Based Syst. (2011)
Keyphrases
- spam filtering
- hypothesis testing
- feature selection algorithms
- feature selection
- text classification
- data sets
- selection algorithm
- feature subset
- statistical tests
- feature set
- neural network
- probability distribution
- confidence intervals
- active learning
- spam filters
- hypothesis tests
- worst case
- information extraction
- reinforcement learning