Semi-Synthetic Data for Enhanced SMS Spam Detection: [Using Synthetic Minority Oversampling TEchnique SMOTE].
Ala EshmawiSuku NairPublished in: MEDES (2014)
Keyphrases
- synthetic data
- spam detection
- minority class
- class imbalance
- class distribution
- class imbalanced
- majority class
- real world
- classification error
- imbalanced datasets
- nearest neighbour
- imbalanced data
- web spam
- support vector machine
- data sets
- cost sensitive learning
- original data
- decision boundary
- spam filtering
- cost sensitive
- fraud detection
- real image data
- sampling methods
- active learning
- training data
- training dataset
- training set
- web graph
- ensemble learning
- web spam detection
- base classifiers
- high dimensionality
- knowledge discovery
- information extraction