增加新闻流量监测板块
This commit is contained in:
oficcejo
@@ -0,0 +1,702 @@
|
||||
"""
|
||||
新闻流量数据获取模块
|
||||
用于获取各大平台的热点新闻和流量数据
|
||||
支持22个平台，包含排名、K值计算等功能
|
||||
"""
|
||||
import requests
|
||||
import logging
|
||||
from datetime import datetime, timedelta
|
||||
from typing import Dict, List, Optional, Tuple
|
||||
import time
|
||||
from collections import Counter
|
||||
|
||||
logging.basicConfig(level=logging.INFO)
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
||||
class NewsFlowDataFetcher:
|
||||
"""新闻流量数据获取器"""
|
||||
|
||||
def __init__(self):
|
||||
self.base_url = "https://newsapi.ws4.cn/api/v1/dailynews/"
|
||||
self.timeout = 10
|
||||
|
||||
# 支持的平台配置 - 扩展到22个平台
|
||||
self.platforms = {
|
||||
# 社交媒体平台（核心流量指标）- 8个
|
||||
'weibo': {'name': '微博热搜', 'category': 'social', 'weight': 10, 'influence': 'high'},
|
||||
'douyin': {'name': '抖音热点', 'category': 'social', 'weight': 9, 'influence': 'high'},
|
||||
'zhihu': {'name': '知乎热榜', 'category': 'social', 'weight': 7, 'influence': 'medium'},
|
||||
'bilibili': {'name': '哔哩哔哩', 'category': 'social', 'weight': 6, 'influence': 'medium'},
|
||||
'xiaohongshu': {'name': '小红书', 'category': 'social', 'weight': 7, 'influence': 'medium'},
|
||||
'kuaishou': {'name': '快手', 'category': 'social', 'weight': 6, 'influence': 'medium'},
|
||||
'tieba': {'name': '百度贴吧', 'category': 'social', 'weight': 5, 'influence': 'low'},
|
||||
'weixin': {'name': '微信热点', 'category': 'social', 'weight': 8, 'influence': 'high'},
|
||||
|
||||
# 新闻媒体平台 - 6个
|
||||
'baidu': {'name': '百度热搜', 'category': 'news', 'weight': 8, 'influence': 'high'},
|
||||
'jinritoutiao': {'name': '今日头条', 'category': 'news', 'weight': 7, 'influence': 'high'},
|
||||
'tenxunwang': {'name': '腾讯网', 'category': 'news', 'weight': 6, 'influence': 'medium'},
|
||||
'netease': {'name': '网易新闻', 'category': 'news', 'weight': 6, 'influence': 'medium'},
|
||||
'ifeng': {'name': '凤凰网', 'category': 'news', 'weight': 5, 'influence': 'medium'},
|
||||
'sina': {'name': '新浪新闻', 'category': 'news', 'weight': 6, 'influence': 'medium'},
|
||||
|
||||
# 财经平台（股市相关）- 5个
|
||||
'sina_finance': {'name': '新浪财经', 'category': 'finance', 'weight': 9, 'influence': 'high'},
|
||||
'eastmoney': {'name': '东方财富', 'category': 'finance', 'weight': 9, 'influence': 'high'},
|
||||
'xueqiu': {'name': '雪球', 'category': 'finance', 'weight': 8, 'influence': 'high'},
|
||||
'cls': {'name': '财联社', 'category': 'finance', 'weight': 8, 'influence': 'high'},
|
||||
'wallstreetcn': {'name': '华尔街见闻', 'category': 'finance', 'weight': 7, 'influence': 'medium'},
|
||||
|
||||
# 科技平台 - 3个
|
||||
'tskr': {'name': '36氪', 'category': 'tech', 'weight': 6, 'influence': 'medium'},
|
||||
'sspai': {'name': '少数派', 'category': 'tech', 'weight': 5, 'influence': 'low'},
|
||||
'juejin': {'name': '掘金', 'category': 'tech', 'weight': 5, 'influence': 'low'},
|
||||
}
|
||||
|
||||
# 平台类别权重（用于转化率计算）
|
||||
self.category_weights = {
|
||||
'finance': 1.5, # 财经平台转化率高
|
||||
'social': 1.2, # 社交媒体传播快
|
||||
'news': 1.0, # 新闻媒体正常
|
||||
'tech': 0.8, # 科技平台相关性低
|
||||
}
|
||||
|
||||
# 停用词（过滤无意义的词）
|
||||
self.stop_words = {
|
||||
'的', '是', '在', '了', '和', '与', '等', '为', '将', '被',
|
||||
'有', '一', '个', '上', '下', '中', '大', '新', '年', '月', '日',
|
||||
'这', '那', '其', '之', '也', '要', '就', '不', '我', '你', '他',
|
||||
'来', '去', '到', '说', '会', '能', '都', '对', '着', '让',
|
||||
'从', '以', '及', '或', '如', '还', '没', '很', '更', '最',
|
||||
}
|
||||
|
||||
def get_platform_news(self, platform: str) -> Dict:
|
||||
"""
|
||||
获取单个平台的新闻数据
|
||||
|
||||
Args:
|
||||
platform: 平台代码（如 'weibo', 'baidu'等）
|
||||
|
||||
Returns:
|
||||
{
|
||||
'success': bool,
|
||||
'platform': str,
|
||||
'platform_name': str,
|
||||
'category': str,
|
||||
'weight': int,
|
||||
'influence': str,
|
||||
'data': List[Dict],
|
||||
'count': int,
|
||||
'fetch_time': str,
|
||||
'error': str (如果失败)
|
||||
}
|
||||
"""
|
||||
try:
|
||||
url = f"{self.base_url}?platform={platform}"
|
||||
|
||||
logger.info(f"正在获取 {platform} 平台数据...")
|
||||
response = requests.get(url, timeout=self.timeout)
|
||||
response.raise_for_status()
|
||||
|
||||
data = response.json()
|
||||
|
||||
if data.get('status') == '200':
|
||||
news_list = data.get('data', [])
|
||||
platform_info = self.platforms.get(platform, {})
|
||||
|
||||
# 为每条新闻添加排名信息
|
||||
for i, news in enumerate(news_list):
|
||||
news['rank'] = i + 1
|
||||
news['platform'] = platform
|
||||
|
||||
return {
|
||||
'success': True,
|
||||
'platform': platform,
|
||||
'platform_name': platform_info.get('name', platform),
|
||||
'category': platform_info.get('category', 'other'),
|
||||
'weight': platform_info.get('weight', 5),
|
||||
'influence': platform_info.get('influence', 'medium'),
|
||||
'data': news_list,
|
||||
'count': len(news_list),
|
||||
'fetch_time': datetime.now().strftime('%Y-%m-%d %H:%M:%S'),
|
||||
}
|
||||
else:
|
||||
return {
|
||||
'success': False,
|
||||
'platform': platform,
|
||||
'error': f"API返回错误: {data.get('msg', '未知错误')}"
|
||||
}
|
||||
|
||||
except requests.exceptions.Timeout:
|
||||
return {
|
||||
'success': False,
|
||||
'platform': platform,
|
||||
'error': f"请求超时（{self.timeout}秒）"
|
||||
}
|
||||
except requests.exceptions.ConnectionError:
|
||||
return {
|
||||
'success': False,
|
||||
'platform': platform,
|
||||
'error': "网络连接失败"
|
||||
}
|
||||
except Exception as e:
|
||||
return {
|
||||
'success': False,
|
||||
'platform': platform,
|
||||
'error': f"获取数据失败: {str(e)}"
|
||||
}
|
||||
|
||||
def get_multi_platform_news(self, platforms: List[str] = None,
|
||||
category: str = None) -> Dict:
|
||||
"""
|
||||
获取多个平台的新闻数据
|
||||
|
||||
Args:
|
||||
platforms: 平台列表，None表示获取所有
|
||||
category: 按类别筛选（'social', 'news', 'finance', 'tech'）
|
||||
|
||||
Returns:
|
||||
{
|
||||
'success': bool,
|
||||
'total_platforms': int,
|
||||
'success_count': int,
|
||||
'failed_count': int,
|
||||
'platforms_data': List[Dict],
|
||||
'fetch_time': str
|
||||
}
|
||||
"""
|
||||
# 确定要获取的平台列表
|
||||
if platforms is None:
|
||||
if category:
|
||||
target_platforms = [
|
||||
p for p, info in self.platforms.items()
|
||||
if info.get('category') == category
|
||||
]
|
||||
else:
|
||||
target_platforms = list(self.platforms.keys())
|
||||
else:
|
||||
target_platforms = platforms
|
||||
|
||||
results = []
|
||||
success_count = 0
|
||||
failed_count = 0
|
||||
|
||||
for platform in target_platforms:
|
||||
result = self.get_platform_news(platform)
|
||||
results.append(result)
|
||||
|
||||
if result['success']:
|
||||
success_count += 1
|
||||
else:
|
||||
failed_count += 1
|
||||
|
||||
# 避免请求过快，休息0.3秒
|
||||
time.sleep(0.3)
|
||||
|
||||
return {
|
||||
'success': success_count > 0,
|
||||
'total_platforms': len(target_platforms),
|
||||
'success_count': success_count,
|
||||
'failed_count': failed_count,
|
||||
'platforms_data': results,
|
||||
'fetch_time': datetime.now().strftime('%Y-%m-%d %H:%M:%S'),
|
||||
}
|
||||
|
||||
def extract_stock_related_news(self, platforms_data: List[Dict],
|
||||
keywords: List[str] = None) -> List[Dict]:
|
||||
"""
|
||||
从新闻数据中提取股票相关的新闻
|
||||
|
||||
Args:
|
||||
platforms_data: 平台数据列表
|
||||
keywords: 股票相关关键词列表
|
||||
|
||||
Returns:
|
||||
List[Dict]: 股票相关新闻列表
|
||||
"""
|
||||
if keywords is None:
|
||||
keywords = [
|
||||
# 股市基础词汇
|
||||
'股', '股市', '股票', 'A股', '港股', '美股', '创业板', '科创板', '北交所',
|
||||
'涨停', '跌停', '大涨', '暴涨', '飙升', '暴跌', '涨幅', '跌幅', '翻倍',
|
||||
'概念股', '龙头股', '妖股', '题材股', '白马股', '蓝筹股', '成长股',
|
||||
'上市', 'IPO', '重组', '并购', '收购', '增发', '回购', '减持', '增持',
|
||||
'业绩', '财报', '利好', '利空', '预增', '预减', '盈利', '亏损',
|
||||
'牛市', '熊市', '反弹', '回调', '震荡', '突破', '新高',
|
||||
'主力', '游资', '北向资金', '外资', '机构', '资金流入', '资金流出',
|
||||
'板块', '行业', '赛道', '题材', '轮动', '热点',
|
||||
|
||||
# 热门板块关键词
|
||||
'芯片', '半导体', '光刻机', '封装', '存储',
|
||||
'新能源', '锂电', '光伏', '储能', '风电', '氢能',
|
||||
'AI', '人工智能', '大模型', 'ChatGPT', 'DeepSeek', '算力', 'GPU',
|
||||
'机器人', '人形机器人', '工业机器人', '减速器', '伺服电机',
|
||||
'低空经济', '无人机', 'eVTOL', '飞行汽车',
|
||||
'数据要素', '数字经济', '信创', '国产替代',
|
||||
'医药', '创新药', 'CXO', '医疗器械', '中药',
|
||||
'消费', '白酒', '食品', '旅游', '免税',
|
||||
'军工', '国防', '航空', '航天', '船舶',
|
||||
'汽车', '新能源车', '智能驾驶', '无人驾驶', '充电桩',
|
||||
'地产', '房地产', '楼市', '房价',
|
||||
'金融', '银行', '保险', '券商', '证券',
|
||||
|
||||
# 政策相关
|
||||
'政策', '利率', '降息', '降准', '货币政策', '财政政策',
|
||||
'国常会', '证监会', '央行', '发改委', '工信部',
|
||||
]
|
||||
|
||||
stock_related = []
|
||||
|
||||
for platform_data in platforms_data:
|
||||
if not platform_data.get('success'):
|
||||
continue
|
||||
|
||||
platform = platform_data['platform']
|
||||
platform_name = platform_data['platform_name']
|
||||
category = platform_data['category']
|
||||
weight = platform_data['weight']
|
||||
influence = platform_data.get('influence', 'medium')
|
||||
|
||||
for news in platform_data.get('data', []):
|
||||
title = news.get('title') or ''
|
||||
content = news.get('content') or ''
|
||||
rank = news.get('rank', 99)
|
||||
|
||||
# 检查是否包含股票关键词
|
||||
text = f"{title} {content}"
|
||||
matched_keywords = [kw for kw in keywords if kw in text]
|
||||
|
||||
if matched_keywords:
|
||||
# 计算综合得分（排名越靠前、权重越高得分越高）
|
||||
rank_score = max(0, 100 - rank * 2) # 排名1得98分，排名50得0分
|
||||
weight_score = weight * 10
|
||||
keyword_score = len(matched_keywords) * 5
|
||||
total_score = rank_score + weight_score + keyword_score
|
||||
|
||||
stock_related.append({
|
||||
'platform': platform,
|
||||
'platform_name': platform_name,
|
||||
'category': category,
|
||||
'weight': weight,
|
||||
'influence': influence,
|
||||
'rank': rank,
|
||||
'title': title,
|
||||
'content': content,
|
||||
'url': news.get('url') or '',
|
||||
'source': news.get('source') or platform_name,
|
||||
'publish_time': news.get('publish_time') or '',
|
||||
'matched_keywords': matched_keywords,
|
||||
'keyword_count': len(matched_keywords),
|
||||
'score': total_score,
|
||||
})
|
||||
|
||||
# 按综合得分排序
|
||||
stock_related.sort(key=lambda x: x['score'], reverse=True)
|
||||
|
||||
return stock_related
|
||||
|
||||
def calculate_flow_score(self, platforms_data: List[Dict]) -> Dict:
|
||||
"""
|
||||
计算流量得分
|
||||
|
||||
根据"流量为王"理念，评估当前市场热度
|
||||
|
||||
Returns:
|
||||
{
|
||||
'total_score': int, # 总流量得分（0-1000）
|
||||
'social_score': int, # 社交媒体流量得分
|
||||
'news_score': int, # 新闻媒体流量得分
|
||||
'finance_score': int, # 财经平台流量得分
|
||||
'tech_score': int, # 科技平台流量得分
|
||||
'level': str, # 流量等级（低/中/高/极高）
|
||||
'analysis': str, # 流量分析
|
||||
'platform_details': List[Dict], # 各平台详情
|
||||
}
|
||||
"""
|
||||
scores = {
|
||||
'social': 0,
|
||||
'news': 0,
|
||||
'finance': 0,
|
||||
'tech': 0,
|
||||
}
|
||||
|
||||
platform_details = []
|
||||
|
||||
for platform_data in platforms_data:
|
||||
if not platform_data.get('success'):
|
||||
continue
|
||||
|
||||
category = platform_data['category']
|
||||
weight = platform_data['weight']
|
||||
count = platform_data['count']
|
||||
platform_name = platform_data['platform_name']
|
||||
|
||||
# 计算得分：权重 × 新闻数量
|
||||
score = weight * count
|
||||
scores[category] = scores.get(category, 0) + score
|
||||
|
||||
platform_details.append({
|
||||
'platform': platform_data['platform'],
|
||||
'platform_name': platform_name,
|
||||
'category': category,
|
||||
'count': count,
|
||||
'score': score,
|
||||
})
|
||||
|
||||
# 计算总分
|
||||
total_score = sum(scores.values())
|
||||
|
||||
# 归一化到0-1000
|
||||
if total_score > 0:
|
||||
normalized_score = min(int(total_score / 50), 1000)
|
||||
else:
|
||||
normalized_score = 0
|
||||
|
||||
# 确定流量等级
|
||||
if normalized_score >= 800:
|
||||
level = "极高"
|
||||
analysis = "流量爆发！市场情绪极度活跃，大量新闻热点，存在热点题材炒作机会。建议：密切关注龙头股，注意追高风险。"
|
||||
elif normalized_score >= 500:
|
||||
level = "高"
|
||||
analysis = "流量较高。市场有明确热点，资金活跃度较好。建议：关注热点板块，注意节奏把握。"
|
||||
elif normalized_score >= 200:
|
||||
level = "中"
|
||||
analysis = "流量正常。市场处于常态，有一定热点但不突出。建议：观望为主，等待明确信号。"
|
||||
else:
|
||||
level = "低"
|
||||
analysis = "流量较低。市场情绪低迷，缺乏热点。建议：控制仓位，等待市场转暖。"
|
||||
|
||||
return {
|
||||
'total_score': normalized_score,
|
||||
'social_score': scores.get('social', 0),
|
||||
'news_score': scores.get('news', 0),
|
||||
'finance_score': scores.get('finance', 0),
|
||||
'tech_score': scores.get('tech', 0),
|
||||
'level': level,
|
||||
'analysis': analysis,
|
||||
'platform_details': platform_details,
|
||||
}
|
||||
|
||||
def get_hot_topics(self, platforms_data: List[Dict], top_n: int = 20) -> List[Dict]:
|
||||
"""
|
||||
获取热门话题（基于标题词频分析）
|
||||
|
||||
Returns:
|
||||
List[Dict]: 热门话题列表
|
||||
"""
|
||||
import jieba
|
||||
|
||||
# 收集所有标题及其来源信息
|
||||
all_titles = []
|
||||
title_sources = {} # 记录每个标题来自哪些平台
|
||||
|
||||
for platform_data in platforms_data:
|
||||
if platform_data.get('success'):
|
||||
platform_name = platform_data['platform_name']
|
||||
for news in platform_data.get('data', []):
|
||||
title = news.get('title') or ''
|
||||
if title:
|
||||
all_titles.append(title)
|
||||
if title not in title_sources:
|
||||
title_sources[title] = []
|
||||
title_sources[title].append(platform_name)
|
||||
|
||||
# 分词并统计
|
||||
word_counter = Counter()
|
||||
word_sources = {} # 记录每个词出现在哪些平台
|
||||
|
||||
for title in all_titles:
|
||||
if title:
|
||||
words = jieba.cut(title)
|
||||
for word in words:
|
||||
if len(word) >= 2 and word not in self.stop_words:
|
||||
word_counter[word] += 1
|
||||
if word not in word_sources:
|
||||
word_sources[word] = set()
|
||||
for source in title_sources.get(title, []):
|
||||
word_sources[word].add(source)
|
||||
|
||||
# 获取TOP N
|
||||
hot_topics = []
|
||||
total_titles = len(all_titles) if all_titles else 1
|
||||
|
||||
for word, count in word_counter.most_common(top_n):
|
||||
sources = list(word_sources.get(word, []))
|
||||
cross_platform = len(sources) # 跨平台数
|
||||
heat = min(int(count / total_titles * 1000), 100)
|
||||
|
||||
# 跨平台加成
|
||||
if cross_platform >= 5:
|
||||
heat = min(heat + 20, 100)
|
||||
elif cross_platform >= 3:
|
||||
heat = min(heat + 10, 100)
|
||||
|
||||
hot_topics.append({
|
||||
'topic': word,
|
||||
'count': count,
|
||||
'heat': heat,
|
||||
'cross_platform': cross_platform,
|
||||
'sources': sources[:5], # 最多显示5个来源
|
||||
})
|
||||
|
||||
return hot_topics
|
||||
|
||||
def get_platform_ranking(self, platforms_data: List[Dict]) -> List[Dict]:
|
||||
"""
|
||||
获取跨平台热度排名
|
||||
|
||||
合并所有平台的新闻，按热度排序
|
||||
|
||||
Returns:
|
||||
List[Dict]: 排名列表
|
||||
"""
|
||||
all_news = []
|
||||
|
||||
for platform_data in platforms_data:
|
||||
if not platform_data.get('success'):
|
||||
continue
|
||||
|
||||
platform = platform_data['platform']
|
||||
platform_name = platform_data['platform_name']
|
||||
weight = platform_data['weight']
|
||||
|
||||
for news in platform_data.get('data', []):
|
||||
title = news.get('title') or ''
|
||||
rank = news.get('rank', 99)
|
||||
|
||||
if not title:
|
||||
continue
|
||||
|
||||
# 计算综合热度分数
|
||||
# 排名越靠前分数越高，平台权重越高分数越高
|
||||
heat_score = (100 - rank) * weight
|
||||
|
||||
all_news.append({
|
||||
'title': title,
|
||||
'platform': platform,
|
||||
'platform_name': platform_name,
|
||||
'original_rank': rank,
|
||||
'heat_score': heat_score,
|
||||
'url': news.get('url') or '',
|
||||
'content': news.get('content') or '',
|
||||
})
|
||||
|
||||
# 按热度分数排序
|
||||
all_news.sort(key=lambda x: x['heat_score'], reverse=True)
|
||||
|
||||
# 添加全局排名
|
||||
for i, news in enumerate(all_news):
|
||||
news['global_rank'] = i + 1
|
||||
|
||||
return all_news
|
||||
|
||||
def calculate_viral_coefficient(self, current_data: Dict,
|
||||
previous_data: Dict) -> Dict:
|
||||
"""
|
||||
计算病毒系数K值
|
||||
|
||||
K值 = 当前传播量 / 上期传播量
|
||||
|
||||
Args:
|
||||
current_data: 当前数据
|
||||
previous_data: 上期数据
|
||||
|
||||
Returns:
|
||||
{
|
||||
'k_value': float, # K值
|
||||
'trend': str, # 趋势（指数型/线性/衰减）
|
||||
'analysis': str, # 分析
|
||||
}
|
||||
"""
|
||||
current_score = current_data.get('total_score', 0)
|
||||
previous_score = previous_data.get('total_score', 0)
|
||||
|
||||
if previous_score == 0:
|
||||
k_value = 1.0
|
||||
trend = "无历史数据"
|
||||
analysis = "首次采集，无法计算K值"
|
||||
else:
|
||||
k_value = round(current_score / previous_score, 2)
|
||||
|
||||
if k_value > 1.5:
|
||||
trend = "指数型爆发"
|
||||
analysis = f"K值={k_value}，流量正在指数型增长！这是病毒式传播的特征，题材可能进入加速期。"
|
||||
elif k_value > 1.0:
|
||||
trend = "线性增长"
|
||||
analysis = f"K值={k_value}，流量稳步增长，题材正在发酵中。"
|
||||
elif k_value == 1.0:
|
||||
trend = "平稳"
|
||||
analysis = f"K值={k_value}，流量保持稳定，市场处于平衡状态。"
|
||||
else:
|
||||
trend = "衰减"
|
||||
analysis = f"K值={k_value}，流量正在衰减，题材热度下降，注意风险。"
|
||||
|
||||
return {
|
||||
'k_value': k_value,
|
||||
'current_score': current_score,
|
||||
'previous_score': previous_score,
|
||||
'trend': trend,
|
||||
'analysis': analysis,
|
||||
}
|
||||
|
||||
def detect_flow_type(self, history_scores: List[int],
|
||||
current_score: int) -> Dict:
|
||||
"""
|
||||
识别流量类型（存量流量型 vs 增量流量型）
|
||||
|
||||
存量流量型：出生自带顶流，流量快速到位（政策/大事件）
|
||||
增量流量型：初始流量小，具备病毒传播能力（话题发酵）
|
||||
|
||||
Args:
|
||||
history_scores: 历史得分列表（从旧到新）
|
||||
current_score: 当前得分
|
||||
|
||||
Returns:
|
||||
{
|
||||
'flow_type': str, # 存量流量型/增量流量型/未知
|
||||
'characteristics': List[str], # 特征
|
||||
'time_window': str, # 时间窗口建议
|
||||
'operation': str, # 操作建议
|
||||
}
|
||||
"""
|
||||
if len(history_scores) < 2:
|
||||
return {
|
||||
'flow_type': '未知',
|
||||
'characteristics': ['历史数据不足'],
|
||||
'time_window': '无法判断',
|
||||
'operation': '继续观察',
|
||||
}
|
||||
|
||||
# 计算初始得分和增长率
|
||||
initial_score = history_scores[0] if history_scores else 0
|
||||
avg_score = sum(history_scores) / len(history_scores)
|
||||
|
||||
# 计算增长趋势
|
||||
growth_rates = []
|
||||
for i in range(1, len(history_scores)):
|
||||
if history_scores[i-1] > 0:
|
||||
rate = (history_scores[i] - history_scores[i-1]) / history_scores[i-1]
|
||||
growth_rates.append(rate)
|
||||
|
||||
avg_growth = sum(growth_rates) / len(growth_rates) if growth_rates else 0
|
||||
|
||||
# 判断流量类型
|
||||
if initial_score >= 500:
|
||||
# 初始热度高 -> 存量流量型
|
||||
flow_type = "存量流量型"
|
||||
characteristics = [
|
||||
"初始热度高",
|
||||
"流量快速到位",
|
||||
"可能与政策/大事件相关",
|
||||
]
|
||||
time_window = "时间窗口短（2-3天）"
|
||||
operation = "快进快出，密切关注热度变化"
|
||||
elif avg_growth > 0.2 and len([r for r in growth_rates if r > 0]) >= 2:
|
||||
# 持续增长 -> 增量流量型
|
||||
flow_type = "增量流量型"
|
||||
characteristics = [
|
||||
"初始热度低",
|
||||
"逐步攀升",
|
||||
"具备病毒传播特征",
|
||||
]
|
||||
time_window = "时间窗口长（5-10天）"
|
||||
operation = "可以埋伏，等待加速"
|
||||
else:
|
||||
flow_type = "常规流量"
|
||||
characteristics = [
|
||||
"热度波动正常",
|
||||
"无明显趋势",
|
||||
]
|
||||
time_window = "无特定窗口"
|
||||
operation = "保持观望"
|
||||
|
||||
return {
|
||||
'flow_type': flow_type,
|
||||
'characteristics': characteristics,
|
||||
'time_window': time_window,
|
||||
'operation': operation,
|
||||
'initial_score': initial_score,
|
||||
'current_score': current_score,
|
||||
'avg_growth': round(avg_growth * 100, 1), # 百分比
|
||||
}
|
||||
|
||||
def get_platform_list(self) -> List[Dict]:
|
||||
"""
|
||||
获取所有支持的平台列表
|
||||
|
||||
Returns:
|
||||
List[Dict]: 平台列表
|
||||
"""
|
||||
result = []
|
||||
for code, info in self.platforms.items():
|
||||
result.append({
|
||||
'code': code,
|
||||
'name': info['name'],
|
||||
'category': info['category'],
|
||||
'weight': info['weight'],
|
||||
'influence': info.get('influence', 'medium'),
|
||||
})
|
||||
|
||||
# 按权重排序
|
||||
result.sort(key=lambda x: x['weight'], reverse=True)
|
||||
return result
|
||||
|
||||
def get_platforms_by_category(self) -> Dict[str, List[str]]:
|
||||
"""
|
||||
按类别获取平台列表
|
||||
|
||||
Returns:
|
||||
Dict[str, List[str]]: 类别 -> 平台代码列表
|
||||
"""
|
||||
categories = {}
|
||||
for code, info in self.platforms.items():
|
||||
category = info['category']
|
||||
if category not in categories:
|
||||
categories[category] = []
|
||||
categories[category].append(code)
|
||||
return categories
|
||||
|
||||
|
||||
# 测试代码
|
||||
if __name__ == "__main__":
|
||||
fetcher = NewsFlowDataFetcher()
|
||||
|
||||
# 测试获取平台列表
|
||||
print("=== 支持的平台列表 ===")
|
||||
platforms = fetcher.get_platform_list()
|
||||
print(f"共支持 {len(platforms)} 个平台：")
|
||||
for p in platforms[:5]:
|
||||
print(f" - {p['name']} ({p['code']}) - 权重:{p['weight']}")
|
||||
|
||||
# 测试获取单个平台
|
||||
print("\n=== 测试获取微博热搜 ===")
|
||||
result = fetcher.get_platform_news('weibo')
|
||||
if result['success']:
|
||||
print(f"✅ 成功获取 {result['count']} 条新闻")
|
||||
print(f"前3条标题：")
|
||||
for news in result['data'][:3]:
|
||||
print(f" {news['rank']}. {news['title']}")
|
||||
else:
|
||||
print(f"❌ 失败: {result['error']}")
|
||||
|
||||
print("\n=== 测试获取财经平台 ===")
|
||||
result = fetcher.get_multi_platform_news(category='finance')
|
||||
print(f"总共 {result['total_platforms']} 个平台")
|
||||
print(f"成功 {result['success_count']} 个")
|
||||
print(f"失败 {result['failed_count']} 个")
|
||||
|
||||
# 提取股票相关新闻
|
||||
stock_news = fetcher.extract_stock_related_news(result['platforms_data'])
|
||||
print(f"\n✅ 提取到 {len(stock_news)} 条股票相关新闻")
|
||||
|
||||
# 计算流量得分
|
||||
flow_score = fetcher.calculate_flow_score(result['platforms_data'])
|
||||
print(f"\n流量得分: {flow_score['total_score']}")
|
||||
print(f"流量等级: {flow_score['level']}")
|
||||
|
||||
# 获取热门话题
|
||||
hot_topics = fetcher.get_hot_topics(result['platforms_data'])
|
||||
print(f"\n热门话题TOP5:")
|
||||
for i, topic in enumerate(hot_topics[:5], 1):
|
||||
print(f" {i}. {topic['topic']} (热度:{topic['heat']}, 跨{topic['cross_platform']}平台)")
|
||||
Reference in New Issue
Block a user