UniMelb_NLP-CORE: Integrating predictions from multiple domains and feature sets for estimating semantic textual similarity.
Spandana GellaBahar SalehiMarco LuiKarl GrieserPaul CookTimothy BaldwinPublished in: *SEM@NAACL-HLT (2013)
Keyphrases
- feature set
- multiple domains
- natural language
- semantic similarity
- semantic features
- semantic relationships
- cross domain
- natural language processing
- feature selection
- feature vectors
- classification accuracy
- feature extraction
- similarity measure
- feature space
- individual features
- wordnet
- semantic information
- keywords
- co occurrence
- metadata
- feature types
- information extraction
- multimedia
- text classification
- text mining
- visual features
- training samples
- machine learning
- image processing
- active learning
- support vector