Keyphrases
- similarity measure
- initial set
- input data
- computationally efficient
- experimental evaluation
- information retrieval
- fully automatic
- detection method
- pairwise
- high precision
- classification method
- synthetic data
- mutual information
- feature set
- prior knowledge
- preprocessing
- computational complexity
- denoising
- text categorization
- detection algorithm
- probability distribution
- probabilistic model
- dynamic programming
- cost function
- optimization method
- weighting scheme
- feature weighting
- method finds