Table 6 Summary of The Methods and Resource Used by All Participating Teams in the Family History Information Extraction Task
From: Family member information extraction via neural sequence labeling models with different tag schemes
Type | Description |
|---|---|
Methodology | CRF, Bidirectional LSTM-CRF, Bidirectional CNN-LSTM-CRF, Pattern |
Word embedding | GloVe: https://nlp.stanford.edu/projects/glove/ word2vec: |
Part-of-speech | NLTK (Natural language toolkit), MedPost |
Ontology/Lexicon | MeSH, Mayo Clinic website and UMLS embedding (https://github.com/clinicalml/embeddings) |