Boosting ICD multi-label classification of health records with contextual embeddings and label-granularity.
Alberto BlancoOlatz Perez-de-ViñaspreAlicia PérezArantza CasillasPublished in: Comput. Methods Programs Biomed. (2020)
Keyphrases
- multi label classification
- health records
- medical records
- multi label
- label space
- electronic health records
- free text
- medical data
- health care
- semi supervised learning
- text categorization
- binary classification
- binary classifiers
- image annotation
- classification algorithm
- patient data
- feature selection
- semi supervised
- cost sensitive
- multi class
- text classification
- graph cuts
- information extraction
- sensitive information
- class labels
- learning algorithm
- dimensionality reduction
- database
- ensemble methods
- image classification
- machine learning
- data sets
- image segmentation
- meta level
- clinical data
- domain knowledge
- base classifiers