# 作者：田丰(FontTian)
# 创建时间:'2017/8/17'
import nltk
from ftools import cnlp
import matplotlib as mpl
mpl.rcParams[u'font.sans-serif'] = [u'KaiTi']
mpl.rcParams[u'font.serif'] = [u'KaiTi']
# mpl.rcParams['axes.unicode_minus'] = False # 解决保存图像是负号'-'显示为方块的问题,或者转换负号为字符串
stopwords_path = u'/home/fonttian/Data/NLP/stopwords/CNENstopwords.txt' # 停用词词表
text_path = u'/home/fonttian/Data/NLP/txt/lz.txt' # 设置要分析的文本路径
fontsTools = cnlp.basis(textPath=text_path,
stopwordsPath=stopwords_path,
ReadText=True)
fontsTools.addUserWords([u'路明非'])
cutAndclearStr = fontsTools.getText()
print(u'nltk 文本分析基本操作')
print(u'nltk.word_tokenize(cutAndclearStr')
print(u'nltk.FreqDist(tokenstr)')
tokenstr = nltk.word_tokenize(cutAndclearStr)
print(u'====== 其他基本指标并绘图 ======')
print(u'下面是nltk的其他类 ')
# - * - coding: utf - 8 -*-
# 作者：田丰(FontTian)
# 创建时间:'2017/8/12'
from os import path
from scipy.misc import imread
from ftools import cnlp
d = path.dirname(__file__)
text_path = d + '/txt/lztest.txt' # 设置要分析的文本路径
stopwords_path = d + '/stopwords/CNENstopwords.txt' # 停用词词表
fontsTools = cnlp.basis(textPath=text_path, stopwordsPath=stopwords_path)
fontsTools.addUserWords([u'路明非'])
font_path = d + '/Fonts/simkai.ttf' # 为worldcloud设置中文字体路径没
back_coloring_path = d + '/img/lz1.jpg' # 设置背景图片路径
imgname1 = d + '/WordCloudDefautColors.png' # 保存的图片名字1(只按照背景图片形状)
imgname2 = d + '/WordCloudColorsByImg.png' # 保存的图片名字2(颜色按照背景图片颜色布局生成)
back_coloring = imread(path.join(d, back_coloring_path)) # 设置词云的形状
# 设置词云属性
wc = fontsTools.wordcloud(
text=None,
image_path=back_coloring_path,
to_file_path=imgname2,
show=True,
font_path=font_path, # 设置字体