Constituents Correspond to Word Sequence Patterns among Sentences with Equivalent Predicate-Argument Structures: Unsupervised Constituency Parsing by Span Matching.
Junjie ChenXiangheng HeDanushka BollegalaYusuke MiyaoPublished in: CoRR (2024)
Keyphrases
- predicate argument structures
- sentence level
- predicate argument
- sequence patterns
- parse tree
- opinion retrieval
- novelty detection
- semantic role labeling
- sentiment analysis
- natural language
- tree kernels
- natural language text
- unsupervised learning
- noun phrases
- query words
- dependency parsing
- dna sequences
- natural language processing
- sentiment classification
- semantic relevance
- relation extraction
- word sense disambiguation
- semantic analysis
- multi document summarization
- n gram
- co occurrence
- structured data
- information extraction
- semi supervised
- semantic representation
- relevance model