|
| 1 | +from jieba import analyse |
| 2 | +from PIL import Image |
| 3 | +from wordcloud import WordCloud, ImageColorGenerator |
| 4 | +import matplotlib.pyplot as plt |
| 5 | +import numpy as np |
| 6 | + |
| 7 | +fenCi = {} |
| 8 | +ciYunArray = [] |
| 9 | +def main(): |
| 10 | + |
| 11 | + # 负责过滤的词语 |
| 12 | + filterWords = ['熟悉', '熟练', '经验', '优先', '应用开发', '相关', '工作', '开发', '能力', '负责', '技术', '具备', '精通', '数据', 'ETC'] |
| 13 | + |
| 14 | + # 结巴分词基于 TF-IDF 算法的关键词 |
| 15 | + tfidf = analyse.extract_tags |
| 16 | + |
| 17 | + for zpInfo in open('sh.txt', 'r', encoding='utf-8'): |
| 18 | + |
| 19 | + if zpInfo.strip() == '': |
| 20 | + continue |
| 21 | + # 详情数据是用&&&分割的 |
| 22 | + infos = zpInfo.split("&&&") |
| 23 | + words = tfidf(infos[-1]) |
| 24 | + |
| 25 | + words = [x.upper() for x in words if x.upper() not in filterWords] |
| 26 | + |
| 27 | + for word in words: |
| 28 | + word = word.upper() |
| 29 | + num = fenCi.get(word, 0) + 1 |
| 30 | + fenCi[word] = num |
| 31 | + |
| 32 | + print(sorted(fenCi.items(), key=lambda kv: (kv[1], kv[0]), reverse=True)) |
| 33 | + print('分出了' + str(len(fenCi)) + '了词语') |
| 34 | + |
| 35 | + |
| 36 | +def getWordCloud(): |
| 37 | + path_img = "python.jpg" |
| 38 | + background_image = np.array(Image.open(path_img)) |
| 39 | + |
| 40 | + wordcloud = WordCloud( |
| 41 | + font_path="/System/Library/Fonts/STHeiti Light.ttc", # 字体 |
| 42 | + background_color="white", |
| 43 | + mask=background_image).generate(" ".join(list(fenCi.keys()))) |
| 44 | + image_colors = ImageColorGenerator(background_image) |
| 45 | + plt.imshow(wordcloud.recolor(color_func=image_colors), interpolation="bilinear") |
| 46 | + plt.axis("off") |
| 47 | + plt.show() |
| 48 | + |
| 49 | + |
| 50 | +if __name__ == '__main__': |
| 51 | + main() |
| 52 | + getWordCloud() |
0 commit comments