-- coding: utf-8 --

""" Created on Mon Oct 29 12:25:50 2018

@author: xyz """

MODELDIR='ltp_data_v3.4.0' ltp = ltp_api(MODELDIR) sentences = SentenceSplitter.split(paragraph)

for sentence in sentences: if sentence: print('\n===================== 原句 =====================\n') print(sentence) # 第一种：类内计算 words = ltp.ltp_segmentor(sentence) # 分词 postags = ltp.ltp_postagger(words) # 词性 arcs = ltp.ltp_parser(words,postags) #依存 netags = ltp.ltp_recognizer(words,postags)# 命名实体识别 labeller = ltp.ltp_labeller(words,postags, arcs) #语义角色

    #ltp.get_result(sentence)
    #output = ltp.output
    #arcs = output['arcs']
    #netags = output['netags']
    #postags = output['postags']
    #labeller = output['role']
    #words = output['words']
    tuples_words = Parser2dataframe(words,postags,arcs)

    print('\n----- 搭配用语查找 -----\n')
    print(FindCollocation(tuples_words))
    print('\n----- 并列词查找 -----\n')
    print(FindSynonym(tuples_words))
    print('\n----- 核心观点抽取 -----\n')
    print(CoreExtraction(tuples_words, words))
    print('\n----- 实体名词搭配 -----\n')
    print(FindEntityCollocation(tuples_words))