diff --git a/scrape_2.py b/scrape_2.py new file mode 100644 index 0000000..b9a7c89 --- /dev/null +++ b/scrape_2.py @@ -0,0 +1,155 @@ +import requests +import re +import time +import pandas as pd +import matplotlib.pyplot as plt +from wordcloud import WordCloud +from collections import Counter + +headers = { + 'cookie':'CURRENT_FNVAL=4048; buvid4=CDB22228-76EA-BC93-F037-78FC6CEC077D36275-023090719-X83v1qigvaVs%2BeTu3%2F5T2g%3D%3D; rpdid=|(u)luk)m|)R0J\'uYmR|)Y)J); enable_web_push=DISABLE; header_theme_version=CLOSE; DedeUserID=567151924; DedeUserID__ckMd5=3d51b3cb3879b2e0; PVID=1; buvid3=909EA327-0037-5349-5CF4-5B2C4EF5300103254infoc; b_nut=1726374703; bsource=search_bing; _uuid=A28E1582-10535-12B8-49B2-D1F1026722D1607876infoc; buvid_fp=1ed033837a881c0f6fee6ce1ae293ed0; bili_ticket=eyJhbGciOiJIUzI1NiIsImtpZCI6InMwMyIsInR5cCI6IkpXVCJ9.eyJleHAiOjE3MjY2MzM5MDQsImlhdCI6MTcyNjM3NDY0NCwicGx0IjotMX0.52ZE1RDG5tSqHh-ZOGgEHDzj6W1UyOtiMkcxw_a2WNY; bili_ticket_expires=1726633844; SESSDATA=701189c6%2C1741926705%2C58e19%2A91CjCANFin8L-nRK6CjxH9_BSgRe6HHUSWybFilZklu8yORRObfCV2cnJswJPECKKy1UcSVkMtcE4ydHItZF9lOW43ZFpyelRVVEUzZUVCdlh6S2ltWWJIaTg0MU1DclRIeG8wbE84cE1pSFBkOXA1alNxTkp3bDJuLWNCN2IzV2JXX2p4SGIxaW9BIIEC; bili_jct=98cbd1d0535939dc4a5c474a44d27ad7; sid=7wfo5xb3; home_feed_column=5; browser_resolution=1432-776; CURRENT_QUALITY=80; bp_t_offset_567151924=977713175669506048; b_lsid=811052DA4_191FB79926A', + 'Referer': 'https://search.bilibili.com/all?vt=82484714&keyword=2024%E5%B7%B4%E9%BB%8E%E5%A5%A5%E8%BF%90%E4%BC%9A&from_source=webtop_search&spm_id_from=333.1007&search_source=5', + 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/128.0.0.0 Safari/537.36 Edg/128.0.0.0' +} + +def get_response(url): + response = requests.get(url = url, headers = headers) + return response + +def get_bv(page_num): + bv_list = [] + link = 'https://api.bilibili.com/x/web-interface/wbi/search/type' + data = { + 'category_id':'', + 'search_type': 'video', + 'ad_resource': '5654', + '__refresh__': 'true', + '_extra':'', + 'context':'', + 'page': page_num, + 'page_size': '42', + 'pubtime_begin_s': '0', + 'pubtime_end_s': '0', + 'from_source':'', + 'from_spmid': '333.337', + 'platform': 'pc', + 'highlight': '1', + 'single_column': '0', + 'keyword': '2024巴黎奥运会', + 'qv_id': 'Hz50pRYmKFQYlX2AorY3bJUTNJbLRwnX', + 'source_tag': '3', + 'gaia_vtoken':'', + 'dynamic_offset': '30', + 'web_location': '1430654', + 'w_rid': '1b994979977a17ee8010f012d43fa7b6', + 'wts': '1726501056' + } + link_data = requests.get(url = link, headers = headers, params = data).json() + for index in link_data['data']['result']: + bv_list.append(index['bvid']) +# pprint(bv_list) + return bv_list + +def get_cid(bvid): + url = f"https://api.bilibili.com/x/web-interface/view?bvid={bvid}" + data = get_response(url).json() +# pprint(data) + return data['data']['cid'] + +def get_danmaku(cid): + url = f"https://comment.bilibili.com/{cid}.xml" + response = get_response(url) + response.encoding = 'utf-8' +# pprint(response.text) + return response.text + +def parse_danmaku(danmaku_xml): + danmaku_list = re.findall('">(.*?)',danmaku_xml) + return danmaku_list + +# 弹幕相关关键词 +ai_keywords = ["AI", "人工智能", "机器学习", "深度学习", "AI技术", "自动驾驶", "智能", "图像识别", "AI应用","智造"] + +# 筛选与AI相关的弹幕 +def get_ai_danmaku(danmaku_list_all, ai_keywords): + ai_related_danmaku = [danmaku for danmaku in danmaku_list_all if any(keyword in danmaku for keyword in ai_keywords)] + return ai_related_danmaku + +# 统计弹幕出现频次 +def count_danmaku(danmaku_list): + danmaku_counter = Counter(danmaku_list) + return danmaku_counter + +# 获取前n个弹幕及其出现次数 +def get_top_n_danmaku(danmaku_counter, n=8): + return danmaku_counter.most_common(n) + +# 写入 Excel +def write_to_excel(data, filename='danmaku_AI_top8.xlsx'): + df = pd.DataFrame(data, columns=['弹幕内容', '出现次数']) + df.to_excel(filename, index=False) + +# 生成并显示词云图 +def get_wordcloud(danmaku_counter): + wordcloud = WordCloud( + width=800, # 宽度 + height=400, # 高度 + background_color='white', # 背景色 + max_words=100, # 显示的最大词语数量 + colormap='viridis', # 颜色映射 + font_path='msyh.ttc' # 指定字体路径，适应中文显示 + ).generate_from_frequencies(danmaku_counter) + plt.figure(figsize=(10, 5)) # 图像大小 + plt.imshow(wordcloud, interpolation="bilinear") + plt.axis("off") # 关闭坐标轴显示 + plt.show() + +if __name__ == '__main__': + + # 收集前300个视频的bvid + videos = [] + page_num = 1 + while len(videos) < 300: + bv_list = get_bv(str(page_num)) + videos.extend(bv_list) + page_num += 1 + time.sleep(1) # 防止请求过于频繁 + + # 仅保留前300个视频号 + videos = videos[:300] + + all_danmaku = [] + all_ai_related_danmaku = [] + + # 获取每个视频的弹幕并打印 + for bvid in videos: + try: + cid = get_cid(bvid) + danmaku_xml = get_danmaku(cid) + danmaku_list = parse_danmaku(danmaku_xml) + + # 筛选AI相关弹幕 + ai_related_danmaku = get_ai_danmaku(danmaku_list, ai_keywords) + all_ai_related_danmaku.extend(ai_related_danmaku) + + for danmaku in danmaku_list: + all_danmaku.append(danmaku) + except Exception as e: + print(f"Error fetching danmaku for video {bvid}: {e}") + time.sleep(1) # 防止请求过于频繁 + + # 统计ai弹幕出现频次 + ai_danmaku_counter = count_danmaku(all_ai_related_danmaku) + + # 获取前8个频次最高的弹幕 + top_8_danmaku = get_top_n_danmaku(ai_danmaku_counter, n=8) + for dm,cnt in top_8_danmaku: + print(dm) + + # 输出到Excel + write_to_excel(top_8_danmaku) + + # 生成词云 + get_wordcloud(ai_danmaku_counter) + +# print("弹幕统计结果已保存到Excel: danmaku_AI_top8.xlsx")