Linked-DocRED - Enhancing DocRED with Entity-Linking to Evaluate End-To-End Document-Level Information Extraction Pipelines.
Pierre-Yves GenestPierre-Edouard PortierElöd Egyed-ZsigmondMartino LovisettoPublished in: SIGIR (2023)
Keyphrases
- end to end
- document level
- information extraction
- language model
- sentence level
- sentiment classification
- query expansion
- natural language processing
- precision and recall
- information retrieval
- relevance model
- pseudo relevance feedback
- semi structured
- machine learning
- text mining
- language modeling
- web documents
- sentiment analysis
- document retrieval
- machine translation
- structured data
- knowledge representation
- named entities
- relational databases
- question answering
- text documents
- probabilistic model
- text summarization
- metadata