Pandas和Pyecharts带你揭秘最近热播好剧的主题和题材趋势

import requests
from bs4 import BeautifulSoup
import pandas as pd
from pyecharts import Bar

# 设置代理信息
proxyHost = "www.16yun.cn"
proxyPort = "5445"
proxyUser = "16QMSOML"
proxyPass = "280651"

# 发送HTTP请求，获取热播剧的页面数据
url = "https://m.iqiyi.com/热播剧"
proxies = {
    "http": f"http://{proxyUser}:{proxyPass}@{proxyHost}:{proxyPort}",
    "https": f"https://{proxyUser}:{proxyPass}@{proxyHost}:{proxyPort}"
}
response = requests.get(url, proxies=proxies)
html = response.text

# 使用BeautifulSoup解析页面
soup = BeautifulSoup(html, "html.parser")
# 提取主题和题材信息
themes = soup.find_all("div", class_="theme")
genres = soup.find_all("div", class_="genre")

# 将数据存储到DataFrame中
data = {"主题": [], "题材": []}
for theme, genre in zip(themes, genres):
    data["主题"].append(theme.text)
    data["题材"].append(genre.text)
df = pd.DataFrame(data)

# 使用Pandas进行数据处理和分析
theme_counts = df["主题"].value_counts()
genre_counts = df["题材"].value_counts()

# 使用Pyecharts进行数据可视化
bar_theme = Bar("热播好剧主题分布")
bar_theme.add("", theme_counts.index, theme_counts.values)
bar_genre = Bar("热播好剧题材分布")
bar_genre.add("", genre_counts.index, genre_counts.values)

# 展示图表
bar_theme.render("theme.html")
bar_genre.render("genre.html")