findSignificantTerms.py

def findSignificantTerms (ids, label):
    body = { "query": { "bool" : { "must" : [ {"term": { "readmit": label} }, { "ids" : {'values' : ids} } ] } }, "aggregations": { "driver_words": { "significant_terms": { "field": "TOKENS", "size": max_sig_terms, chi_square : {}, "background_filter": { "terms": { "readmit": [ 0, 1 ] } } } } }, "size": 0 }

    response = client.search(index=index,body=body, request_timeout=6000)
    max_score = response['aggregations']['driver_words']['buckets'][0]['score']
    words = [bucket['key'] for bucket in response['aggregations']['driver_words']['buckets']]
    if (boosting == 'yes'):
        boosts = [bucket['score']/max_score for bucket in response['aggregations']['driver_words']['buckets']]
    else:
        boosts = [1.0] * len(words)
    return words, boosts