scripts-for-PhenoSSU

instruction for list of files

1. attribute_trigger_terms_from_training_set.xlsx
   the file records triggers terms of attributes arranged from the training set.

2. diseases_for_training.txt and diseases_for_test.txt
   the file records names of documents used for training and testing.

3. wiki_annotation_normalized_and_reviewed.xlsx
   the file records the normalization process of wiki annotations to SNOMED-CT concepts.

4. metamap annotation and phenoSSU modelling.ipynb
   the file records the process of annotating corpus with metamap; and the process of modelling PhenoSSU with co-occurence based method.

5. building and evaluating SVM-based models.ipynb
   the file records the process of training and evalution of SVM-based attribute value prediction models.

6. building and evaluating BiLSTM-based models.ipynb
   the file records the process of training and evalution of BiLSTM-based attribute value prediction models.