A graph-based approach for positive and unlabeled learning

Computer Science

• k-Nearest Neighbors Algorithm 50%
• Labeled Document 100%
• Similarity Graph 50%
• Text Collection 50%
• Text Information 50%
• Vector Space Models 100%

Keyphrases

• K-nearest Neighbor (K-NN) 14%
• Label Propagation 28%
• Positive Learning 100%
• Similarity Graph 14%
• Text Collection 14%
• Text Information 14%
• Unlabeled Learning 100%
• Vector Space Model 28%