为了编写一个Python程序来爬取B站视频的评论并做情感分析，你需要遵循以下几个步骤： 首先，安装必要的库： ```bash pip install requests beautifulsoup4 pandas nltk textblob ``` 1. 导入库并设置请求头（模拟浏览器访问以避免被屏蔽）： ```python import requests from bs4 import BeautifulSoup import pandas as pd from textblob import TextBlob # 设置User-Agent headers = { 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.3' } ``` 2. 爬取B站评论： ```python def get_comments(url): response = requests.get(url, headers=headers) soup = BeautifulSoup(response.text, 'lxml') comments_div = soup.find('div', {'class': 'comment-item-list'}) comment_list = [] for comment in comments_div.find_all('li'): raw_comment = comment.find('span', {'class': 'content'}).text comment_list.append(raw_comment) return comment_list url = "https://www.bilibili.com/video/BV某个视频ID评论页" comments = get_comments(url) ``` 3. 将评论数据处理成DataFrame： ```python data = [{'comment': c} for c in comments] df = pd.DataFrame(data) ``` 4. 清洗文本、去除无关字符和标点符号，并分词： ```python df['cleaned_text'] = df['comment'].apply(lambda x: " ".join(x.strip().split())) ``` 5. 情感分析（TextBlob库）： ```python def analyze_sentiment(text): analysis = TextBlob(text) sentiment = analysis.sentiment.polarity if sentiment > 0: return '正向' elif sentiment < 0: return '负向' else: return '中立' df['sentiment'] = df['cleaned_text'].apply(analyze_sentiment) ``` 6. 结果汇总： ```python result_summary = df.groupby('sentiment').size() print(result_summary) ```