英文文本关键词抽取——使用NLTK进行关键词抽取_使用nltk抽取 英语文本关键词 例子

"""
__author__:shuangrui Guo
__description__:
"""
import sys
import nltk
import json
from tqdm import tqdm
#多进程的包
import multiprocessing
import argparse
import os
import re
SUFFIX_NLTK = '__nltk.json'
#清洗文本
def clean_text(text):
    text = re.sub(r'[^\x00-\x7F]+',' ',text)
    text = re.sub(r"([.,!:?()])",r" \1 ",text)
    text = re.sub(r"\s{2,}"," ",text)
    text = text.replace("-"," ")
    return text
#获取文件行数的函数
def get_line_count(inFile):
    lines = 0
    with open(inFile,'r') as f:
        while f.readline():
            lines+=1
    return lines
#跳过所有的单个词，默认是True
def get_nps_from_tree(tree, words_original, attachNP=False, skip_single_word=True):
    nps = []
    st = 0
    for subtree in tree:
        if isinstance(subtree, nltk.tree.Tree):
            if subtree.label() == 'NP':
                np = subtree.leaves()
                ed = st + len(np)
                if not skip_single_word or len(np) > 1:
                    nps.append({'st': st,
                                'ed': ed,
                                'text': ' '.join(words_original[st:ed])})
                    if attachNP:
                        nps[-1]['np'] = np
            st += len(subtree.leaves())
        else:
            st += 1
    return nps
def validate_nps(nps, words_original):
    validated_nps = []
    for np in sorted(nps, key=lambda x:x['st']):
        st = np['st']
        ed = np['ed']
        token_span = words_original[st:ed]
        # 'A polynomial time algorithm for the Lambek calculus with brackets of  bounded order'
        if ' '.join(token_span).strip() != np['text'].strip():
            print(' '.join(token_span))
            print(np)
            return validated_nps
        validated_nps.append(np)
    return nps
def get_nps_nltk_raw(doc):
    # 预先定义的分块语法，具体含义不清楚
    GRAMMAR = r"""
    NBAR:
      {<NN.*|JJ>*<NN.*>}  # Nouns and Adjectives, terminated with Nouns（名次和形容词，并且以名词结尾）
    NP:
      {<NBAR>}
      {<NBAR><IN><NBAR>}  # Above, connected with in/of/etc...
    """
    # 定义语法解析器
    _PARSER = nltk.RegexpParser(GRAMMAR)
    doc = clean_text(doc)
    #对文档使用空格切分
    words_original = nltk.word_tokenize(doc)
    #words_original = doc.split(' ')
    try:
        parse_tree = _PARSER.parse(nltk.pos_tag(words_original))
        pass
    except Exception as e:
        import ipdb; ipdb.set_trace()
        pass
    nps = get_nps_from_tree(parse_tree, words_original)
    return nps
#读入与写出文件
def writeToJson(inFile, outFile):
    #分别读入文件，与写出文件
    with open(inFile, 'r') as fin, open(outFile, 'w') as fout:
        total = get_line_count(inFile)
        for line in tqdm(fin, total=total):
            doc = line.strip('\r\n')
            #对每一行进行处理
            if doc:
                nps = get_nps_nltk_raw(doc)
            else:
                nps = []
            fout.write(json.dumps(nps))
            fout.write('\n')
 
if __name__ == '__main__':
    inFile = "./patent_abstract.txt"
    outFile = inFile + SUFFIX_NLTK
    writeToJson(inFile, outFile)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102

原文链接 ：https://blog.csdn.net/qq_28790663/article/details/115956521