SciREX: A Challenge Dataset for Document-Level Information Extraction.
Sarthak JainMadeleine van ZuylenHannaneh HajishirziIz BeltagyPublished in: CoRR (2020)
Keyphrases
- document level
- information extraction
- language model
- sentiment classification
- sentence level
- text mining
- natural language processing
- query expansion
- structured data
- information retrieval
- free text
- named entities
- web documents
- document retrieval
- sentiment analysis
- precision and recall
- named entity recognition
- question answering
- learning algorithm
- information retrieval systems
- text summarization
- data mining
- coreference resolution
- relation extraction
- pseudo relevance feedback
- natural language
- automatic extraction
- domain specific
- text documents
- conditional random fields
- text classification