A Method of Extracting and Evaluating Popularity and Unpopularity for Natural Language Expressions.
Kazuhiro MoritaYuki KadoyaEl-Sayed AtlamMasao FuketaShinkaku KashijiJun-ichi AoePublished in: KES (2004)
Keyphrases
- natural language
- high accuracy
- preprocessing
- main contribution
- objective function
- pairwise
- significant improvement
- cost function
- experimental evaluation
- computationally efficient
- model selection
- prior knowledge
- synthetic data
- detection method
- fully automatic
- question answering
- high precision
- decision trees
- optimization algorithm
- mutual information
- optical flow
- feature vectors
- similarity measure
- image sequences