def word_segment():
in_text = codecs.open('data/xuezhong.txt', 'r', encoding='UTF-8').read()
pynlpir.open()
# 添加自定义字典
nlpir.AddUserWord(c_char_p("徐骁".encode()))
nlpir.AddUserWord(c_char_p("老怪物".encode()))
nlpir.AddUserWord(c_char_p("徐渭熊".encode()))
nlpir.AddUserWord(c_char_p("徐北枳".encode()))
nlpir.AddUserWord(c_char_p("白狐儿脸".encode()))
nlpir.AddUserWord(c_char_p("轩辕青锋".encode()))
nlpir.AddUserWord(c_char_p("姜泥".encode()))
nlpir.AddUserWord(c_char_p("大官子".encode()))
nlpir.AddUserWord(c_char_p("北凉".encode()))
nlpir.AddUserWord(c_char_p("小和尚".encode()))
# 对文件分词
nlpir.FileProcess('data/xuezhong.txt'.encode("utf-8"),
'data/xuezhong_seg_1.txt'.encode("utf-8"), False)
# key_words = pynlpir.get_key_words(in_text, max_words=100, weighted=True)
pynlpir.close()
print(key_words)
print("segment finished")
def st_segment():
# 分词
pynlpir.open()
# 添加自定义字典
nlpir.AddUserWord(c_char_p("三体".encode()))
nlpir.AddUserWord(c_char_p("罗辑".encode()))
# 对文件分词
nlpir.FileProcess('data/st.txt'.encode("utf-8"),
'data/segdata/st_seg.txt'.encode("utf-8"), False)
pynlpir.close()
def st_WordCloud():
# 生成三体词云
in_text = codecs.open('data/st.txt', 'r', encoding='UTF-8').read()
pynlpir.open()
nlpir.AddUserWord(c_char_p("三体".encode()))
nlpir.AddUserWord(c_char_p("罗辑".encode()))
key_words = pynlpir.get_key_words(in_text, max_words=300, weighted=True)
# 停用词
stopwords = pd.read_csv("data/stop_words.txt",
index_col=False,
quoting=3,
sep="\n",
names=['stopword'],
encoding='utf-8')
words = [word for word, wegiht in key_words]
keywords_df = pd.DataFrame({'keywords': words})
# 去掉停用词
keywords_df = keywords_df[~keywords_df.keywords.isin(stopwords.stopword.
tolist())]
word_freq = []
for word in keywords_df.keywords.tolist():
for w, k in key_words:
if word == w:
word_freq.append((word, k))
pynlpir.close()
print(word_freq)
font = r'C:\Windows\Fonts\msyh.ttc' # 指定字体，不指定会报错
# color_mask = imread("resource/ge.jpg") # 读取背景图片
color_mask = imread("resource/timg.jpg") # 读取背景图片
wcloud = WordCloud(
font_path=font,
# 背景颜色
background_color="white",
# 词云形状
mask=color_mask,
# 允许最大词汇
max_words=2000,
# 最大号字体
max_font_size=80)
wcloud.generate_from_frequencies(dict(word_freq))
# 以下代码显示图片
plt.imshow(wcloud)
plt.axis("off")
plt.show()
wcloud.to_file("data/wcimage/三体词云_2.png")
def ImDict(fname):
with open(fname) as frd:
for line in frd:
tmp = line.strip()
if tmp:
tmp = tmp.decode("utf-8")
if tmp:
nlpir.AddUserWord(tmp.encode("utf-8"))
def import_AddUserWord(self, word):
nlpir.AddUserWord(str(word))
def add_user_word(path):
# 添加用户词典
[nlpir.AddUserWord(line.strip("\n"))
for line in open(os.path.join(TEXT_RESOURCE, path))]