Python+WordCloud绘制京东评论词云图实践

WordCloud绘制词云图实践案例

作为文本分析，其中常用库jieba、wordcloud等用法演示，以作记录~~

import os
import random
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

import jieba
import jieba.analyse
import wordcloud
from PIL import Image

# 显示中文字体
font_path = '/Library/Fonts/Songti.ttc'

# 设置显示中文所需字体
plt.rcParams['font.family'] = 'Songti SC'
# 设置负号的正常显示
plt.rcParams['axes.unicode_minus'] = False

%matplotlib inline

# 设置清晰度
%config InlineBackend.figure_format = 'retina'

1、数据导入

• 数据集，可从这里下载

# 采集的某品牌商品的最新100条评论
comment_pd = pd.read_excel('.././datas/JD-Comments.xlsx')
comment_pd.info()
# 转换成文本内容
comment_content = ' '.join(comment_pd.comment.to_list())

2、分词处理

jieba, “结巴”中文分词，最好的Python中文分词组件

• 中文停用词表，可从这里查看下载

# 设置停用词库
jieba.analyse.set_stop_words('.././datas/stop_words.txt')

# 基于TextRanks算法关键词提取
comment_words = jieba.analyse.textrank(comment_content, topK=300, withWeight=True)

# 基于TF-IDF算法的关键词提取
comment_words2 = jieba.analyse.extract_tags(comment_content, topK=300, withWeight=True)

# 仅使用关键词，忽略词频
word_list = []
for item in comment_words:
    word_list.append(item[0])
words = ' '.join(word_list)

# 列表数据转换为Dict
frequency_words = {}
for item in comment_words2:
    frequency_words[item[0]] = item[1]

3、展示案例

3.1 默认参数

# 默认参数，最简单例子
wc = wordcloud.WordCloud(font_path=font_path)
wc.generate(words)

# 绘图展示
plt.figure(figsize=(12, 9))
plt.imshow(wc, interpolation='bilinear')
# 关闭XY轴
plt.axis('off')
plt.show()

3.2 设置字体大小

# 默认参数，最简单例子
wc = wordcloud.WordCloud(font_path=font_path, max_font_size=50)
wc.generate(words)

# 绘图展示
plt.figure(figsize=(12, 9))
plt.imshow(wc, interpolation='bilinear')
# 关闭XY轴
plt.axis('off')
plt.show()

3.3 设置图形形状蒙板

• 选择一个图片，取得图片元素形状
• 设置contour_width、contour_color图形轮廓宽度及颜色

# 选择一个图片
mask = np.array(Image.open('.././datas/china-map.jpeg'))
wc = wordcloud.WordCloud(font_path=font_path, background_color='white', max_font_size=50, mask=mask,
                         contour_width=1, contour_color='grey')
wc.generate(words)

# 绘图展示
plt.figure(figsize=(12, 9))
plt.imshow(wc, interpolation='bilinear')
# 关闭XY轴
plt.axis('off')
plt.show()

3.4 使用带词频数据展示

• 方法：generate_from_frequencies

wc = wordcloud.WordCloud(font_path=font_path, background_color='white', max_font_size=100, mask=mask,
                         contour_width=1, contour_color='grey')
wc.generate_from_frequencies(frequency_words)

# 绘图展示
plt.figure(figsize=(12, 9))
plt.imshow(wc, interpolation='bilinear')
# 关闭XY轴
plt.axis('off')
plt.show()

3.5 使用图形配色方案显示词云

• WordCloud.ImageColorGenerator方法
• wc.recolor方法

wc = wordcloud.WordCloud(font_path=font_path, background_color='white', max_font_size=100, mask=mask,
                         contour_width=1, contour_color='grey')
wc.generate_from_frequencies(frequency_words)

# 从图像创建颜色
image_colors = wordcloud.ImageColorGenerator(mask)

# 绘图展示
plt.figure(figsize=(12, 9))
plt.imshow(wc.recolor(color_func=image_colors), interpolation='bilinear')
# 关闭XY轴
plt.axis('off')
plt.show()

3.6 使用自定义颜色展示

def grey_color_func(word, font_size, position, orientation, random_state=None, **kwargs):
    return "hsl(0, 0%%, %d%%)" % random.randint(60, 100)

wc = wordcloud.WordCloud(font_path=font_path, background_color='white', max_words=100, max_font_size=100, 
                         mask=mask, contour_width=1, contour_color='grey')
wc.generate_from_frequencies(frequency_words)

# 默认颜色
default_colors = wc.to_array()

# 绘图展示
plt.figure(figsize=(12, 9))
plt.imshow(wc.recolor(color_func=grey_color_func, random_state=3), interpolation='bilinear')
# 关闭XY轴
plt.axis('off')

# 默认图
plt.figure(figsize=(12, 9))
plt.imshow(default_colors, interpolation='bilinear')
# 关闭XY轴
plt.axis('off')
plt.show()