ner_similarity.py 文件源码

 复制代码def extract_entities(text):
    result=dict()
    for chunk in nltk.ne_chunk(nltk.pos_tag(nltk.word_tokenize(text))):
        # chunk.draw()
        if(isinstance(chunk, nltk.tree.Tree)):
            for subtree in chunk.subtrees(filter=lambda t: (t.label() == 'PERSON' or t.label() == 'GPE' or t.label() == 'LOCATION')):
                for leave in subtree.leaves():
                    if leave[0].lower() not in irrelevant_loc_words:
                        result[leave[0].lower()]=subtree.label()
    # print result
    return result