Question

我想使用Python Stanford NER模块，但一直收到错误，我在互联网上搜索但没有得到任何结果。以下是错误的基本用法。

import ner
tagger = ner.HttpNER(host='localhost', port=8080)
tagger.get_entities("University of California is located in California,   

United States")

错误

Traceback (most recent call last):
File "<pyshell#3>", line 1, in <module>
tagger.get_entities("University of California is located in California, United States")
File "C:\Python27\lib\site-packages\ner\client.py", line 81, in get_entities
tagged_text = self.tag_text(text)
File "C:\Python27\lib\site-packages\ner\client.py", line 165, in tag_text
c.request('POST', self.location, params, headers)
File "C:\Python27\lib\httplib.py", line 1057, in request
self._send_request(method, url, body, headers)
File "C:\Python27\lib\httplib.py", line 1097, in _send_request
self.endheaders(body)
File "C:\Python27\lib\httplib.py", line 1053, in endheaders
self._send_output(message_body)
File "C:\Python27\lib\httplib.py", line 897, in _send_output
self.send(msg)
File "C:\Python27\lib\httplib.py", line 859, in send
self.connect()
File "C:\Python27\lib\httplib.py", line 836, in connect
self.timeout, self.source_address)
File "C:\Python27\lib\socket.py", line 575, in create_connection
raise err
error: [Errno 10061] No connection could be made because the target machine actively refused it

使用安装了最新Java的Windows 10

Answer 1

Python Stanford NER模块是斯坦福NER的包装器允许您运行python命令以使用NER服务。
NER service是Python模块的独立实体。这是一个Java 程序。要通过python或任何其他方式访问此服务，请执行此操作首先需要启动服务。
有关如何启动Java的详细信息程序/服务可以在这里找到 - http://nlp.stanford.edu/software/CRF-NER.shtml
NER自带一个用于windows的.bat文件和一个用于unix / linux的.sh文件。我认为这些文件以GUI
要在没有GUI的情况下启动服务，您应该运行与此类似的命令：
java -mx600m -cp stanford-ner.jar edu.stanford.nlp.ie.crf.CRFClassifier -loadClassifier classifiers/english.all.3class.distsim.crf.ser.gz
这将运行NER jar，设置内存，并设置您要使用的分类器。（我想你必须在斯坦福NER目录中运行它）
一旦NER程序运行，您就可以运行python代码并查询NER。

Answer 2

这是python 3x中完整的Stanford NER脚本

此代码将从“ TextFilestoTest”文件夹中读取每个文本文件，并检测实体并将其存储在数据框中（测试）

import os
import nltk
import pandas as pd
import collections

from nltk.tag import StanfordNERTagger
from nltk.tokenize import word_tokenize


stanford_classifier = 'ner-trained-EvensTrain.ser.gz'
stanford_ner_path = 'stanford-ner.jar'

# Creating Tagger Object
st = StanfordNERTagger(stanford_classifier, stanford_ner_path, encoding='utf-8')

java_path = "C:/Program Files (x86)/Java/jre1.8.0_191/bin/java.exe"
os.environ['JAVAHOME'] = java_path


def get_continuous_chunks(tagged_sent):
    continuous_chunk = []
    current_chunk = []

    for token, tag in tagged_sent:
        if tag != "0":
            current_chunk.append((token, tag))
        else:
            if current_chunk: # if the current chunk is not empty
                continuous_chunk.append(current_chunk)
                current_chunk = []
    # Flush the final current_chunk into the continuous_chunk, if any.
    if current_chunk:
        continuous_chunk.append(current_chunk)
    return continuous_chunk

TestFiles = './TextFilestoTest/'
files_path = os.listdir(TestFiles)    
Test = {}

for i in files_path:
    p = (TestFiles+i)
    g= (os.path.splitext(i)[0])
    Test[str(g)] = open(p, 'r').read()

## Predict labels of all words of 200 text files and inserted into dataframe
df_fin = pd.DataFrame(columns = ["filename","Word","Label"])
for i in Test:
    test_text = Test[i]
    test_text = test_text.replace("\n"," ")
    tokenized_text = test_text.split(" ")
    classified_text = st.tag(tokenized_text)
    ne_tagged_sent = classified_text
    named_entities = get_continuous_chunks(ne_tagged_sent)

    flat_list = [item for sublist in named_entities for item in sublist]

    for fl in flat_list:
        df_ = pd.DataFrame()
        df_["filename"]  = [i]
        df_["Word"]  = [fl[0]]
        df_["Label"]  = [fl[1]]
        df_fin = df_fin.append(df_)

df_fin_vone = pd.DataFrame(columns = ["filename","Word","Label"])
test_files_len = list(set(df_fin['filename']))

如果下面有任何问题发表评论，我会回答。谢谢

无法在python模块中使用Stanford NER

2 个答案: