Do You Need Embeddings Trained on a Massive Specialized Corpus for Your Clinical Natural Language Processing Task?
Antoine NeurazVincent LootenBastien RanceNicolas DanielNicolas GarcelonLeonardo Campillos LlanosAnita BurgunSophie RossetPublished in: MedInfo (2019)
Keyphrases
- natural language processing
- broad coverage
- pos taggers
- information extraction
- general purpose
- reference resolution
- machine learning
- text mining
- natural language
- training set
- pos tagging
- semantic relations
- named entity recognition
- computational biology
- wordnet
- clinical applications
- computational linguistics
- low dimensional
- distance measure
- manually annotated
- coreference resolution
- medical data
- patient data
- free text
- sentiment analysis
- knowledge representation
- linguistic knowledge
- machine translation
- question answering
- high dimensional data
- text processing
- dimensionality reduction
- clinical trials
- clinical practice
- text corpora
- patient records
- medical knowledge
- clinical guidelines
- traditional chinese medicine
- medical treatment
- part of speech
- semantic analysis
- clinical data
- word sense disambiguation
- manifold learning
- multilayer perceptron
- vector space
- test set
- decision support system
- data analysis
- feature selection
- information retrieval