python处理自然语言：1、调用LTP的API，2、使用pyltp

# -*- coding: utf-8 -*-
import urllib2

url_get_base = "http://api.ltp-cloud.com/analysis/?"
api_key = ‘*********‘      # 输入注册API_KEY

# 待分析的文本
text = "国务院总理李克强调研上海外高桥时提出，支持上海积极探索新机制。"

format0 = ‘plain‘  # 结果格式，有xml、json、conll、plain（不可改成大写）
pattern = ‘ws‘  # 指定分析模式，有ws、pos、ner、dp、sdp、srl和all

#分词
result = urllib2.urlopen("%sapi_key=%s&text=%s&format=%s&pattern=%s"
                         % (url_get_base, api_key, text, format0, ‘ws‘))
content = result.read().strip()
print content
print ‘*‘*60
#词性标注
result1 = urllib2.urlopen("%sapi_key=%s&text=%s&format=%s&pattern=%s"
                         % (url_get_base, api_key, text, format0, ‘pos‘))
content1 = result1.read().strip()
print content1
print ‘*‘*60
#命名实体识别
result2 = urllib2.urlopen("%sapi_key=%s&text=%s&format=%s&pattern=%s"
                         % (url_get_base, api_key, text, format0, ‘ner‘))
content2 = result2.read().strip()
print content2
print ‘*‘*60
#依存句法分析
result3 = urllib2.urlopen("%sapi_key=%s&text=%s&format=%s&pattern=%s"
                         % (url_get_base, api_key, text, format0, ‘dp‘))
content3 = result3.read().strip()
print content3
print ‘*‘*60
#语义依存分析
result4 = urllib2.urlopen("%sapi_key=%s&text=%s&format=%s&pattern=%s"
                         % (url_get_base, api_key, text, format0, ‘sdp‘))
content4 = result4.read().strip()
print content4
print ‘*‘*60
#语义角色标注
result1 = urllib2.urlopen("%sapi_key=%s&text=%s&format=%s&pattern=%s"
                         % (url_get_base, api_key, text, format0, ‘srl‘))
content1 = result1.read().strip()
print content1
print ‘*‘*60

# -*- coding: utf-8 -*-
from pyltp import SentenceSplitter
sents = SentenceSplitter.split(‘元芳你怎么看？我就趴窗口上看呗！‘)  # 分句
print ‘\n‘.join(sents)

# -*- coding: utf-8 -*-
import os
LTP_DATA_DIR = ‘/path/to/your/ltp_data‘  # ltp模型目录的路径
cws_model_path = os.path.join(LTP_DATA_DIR, ‘cws.model‘)  # 分词模型路径，模型名称为`cws.model`

from pyltp import Segmentor
segmentor = Segmentor()  # 初始化实例
segmentor.load(cws_model_path)  # 加载模型
words = segmentor.segment(‘元芳你怎么看‘)  # 分词
print ‘\t‘.join(words)
segmentor.release()  # 释放模型

# -*- coding: utf-8 -*-
from pyltp import NamedEntityRecognizer
recognizer = NamedEntityRecognizer() # 初始化实例
recognizer.load(‘/path/to/your/model‘)  # 加载模型
netags = recognizer.recognize(words, postags)  # 命名实体识别
print ‘\t‘.join(netags)
recognizer.release()  # 释放模型

# -*- coding: utf-8 -*-
from pyltp import SementicRoleLabeller
labeller = SementicRoleLabeller() # 初始化实例
labeller.load(‘/path/to/your/model/dir‘)  # 加载模型
roles = labeller.label(words, postags, netags, arcs)  # 语义角色标注
for role in roles:
    print role.index, "".join(
        ["%s:(%d,%d)" % (arg.name, arg.range.start, arg.range.end) for arg in role.arguments])
labeller.release()  # 释放模型