NLP Steps Added to VAERS Study Project

graphext-vaccines-nlp-steps.txt

# Configure English as language of text
make_constant(ds["Symptom Description"], {
    "value": "en",
    "out_type": "category"
}) -> (ds.lang)

# Parse and extract ADJECTIVES from SYMPTOM DESCRIPTION column.
extract_keywords(ds["Symptom Description"], ds.lang, {
    "keywords": {
        "pos_tags": [
            "ADJ"
        ],
        "entities": false,
        "noun_phrases": false
    },
    "extended_language_support": false
}) -> (ds["Symptom Description - Adjectives"])

# Parse and extract NOUNS from SYMPTOM DESCRIPTION column.
extract_keywords(ds["Symptom Description"], ds.lang, {
    "keywords": {
        "pos_tags": [
            "NOUN"
        ],
        "entities": false,
        "noun_phrases": false
    },
    "extended_language_support": false
}) -> (ds["Symptom Description - Nouns"])

# Parse all text columns and extract their ngrams.
extract_ngrams(ds["Symptom Description"],
               ds.lang) -> (ds["Symptom Description - Significant terms"])

extract_ngrams(ds.Allergies,
               ds.lang) -> (ds["Allergies - Significant terms"])

extract_ngrams(ds["Prior Vaccinations"],
               ds.lang) -> (ds["Prior Vaccinations - Significant terms"])

extract_ngrams(ds["Medical History Notes"],
               ds.lang) -> (ds["Medical History Notes - Significant terms"])

extract_ngrams(ds["Other Medication"],
               ds.lang) -> (ds["Other Medication - Significant terms"])

extract_ngrams(ds["Current Illnesses"],
               ds.lang) -> (ds["Current Illnesses - Significant terms"])