使用spacy进行文本相似度处理

import spacy

def check_zh_doc_similarity():
    nlp = spacy.load('zh_core_web_lg')
    doc1 = nlp('你好吗?')
    doc2 = nlp('你还好吗?')
    doc3 = nlp('今天你还好吗?')
    doc4 = nlp('你的身体今天还好吗?')
    print(doc1.similarity(doc2))  # 0.7544851165307768
    print(doc2.similarity(doc3))  # 0.9664107589955437
    print(doc1.similarity(doc3))  # 0.730822854943996
    print(doc1.similarity(doc4))  # 0.6528684500574182

if __name__ == '__main__':
    check_zh_doc_similarity()