Biased Support Vector Machine with Self-Constructed Universum for PU Learning.
Ying ZhangYingjie TianZhiquan QiPublished in: ICDM Workshops (2014)
Keyphrases
- support vector machine
- binary classification
- maximum margin
- unbalanced data
- semi supervised learning
- negative examples
- multi class
- support vector
- misclassification costs
- training data
- svm classifier
- cost sensitive
- instance selection
- feature selection
- hyperplane
- machine learning
- kernel function
- kernel methods
- labeled data
- multiple kernel learning
- semi supervised
- training set
- classification algorithm
- feature space
- support vectors
- high dimensional
- learning algorithm