Interactive Spam Filtering with Active Learning and Feature Selection.
Masayuki OkabeSeiji YamadaPublished in: Web Intelligence/IAT Workshops (2008)
Keyphrases
- spam filtering
- active learning
- feature selection
- text classification
- selective sampling
- text categorization
- machine learning models
- spam filters
- imbalanced data classification
- anti spam
- machine learning
- feature space
- user interaction
- learning strategies
- spam detection
- feature selection algorithms
- semi supervised
- labeled data
- batch mode
- training set
- information gain
- transfer learning
- naive bayes
- unlabeled data
- mutual information
- random sampling
- experimental design
- feature set
- relevance feedback
- feature mapping
- email spam
- feature extraction
- spam classification
- support vector
- databases
- string kernels
- class imbalance
- classification models
- support vector machine
- semi supervised learning
- dimensionality reduction
- supervised learning