#!/usr/bin/env python3 # -*- coding: utf-8 -*- """ 公众情绪分析模块 - 人物画像技能 分析公众讨论的情感倾向和主要话题 创建时间：2026-04-03 开发者：瞰宇 (Kàn Yǔ) """ import json from typing import Dict, List, Tuple, Optional from dataclasses import dataclass from collections import Counter, defaultdict import re @dataclass class EmotionDistribution: """情感分布""" positive: float # 正面比例 neutral: float # 中立比例 negative: float # 负面比例 strong_emotion_ratio: float # 强烈情绪比例 @dataclass class TopicAnalysis: """话题分析""" topic: str mention_frequency: float # 提及频率 sentiment: str # 情感倾向（正面/负面/中立） intensity: str # 情感强度（弱/中/强） @dataclass class SupporterProfile: """支持者画像""" group_name: str demographics: Dict[str, str] # 人口特征 ideology: List[str] # 意识形态特征 emotion_characteristics: List[str] # 情感特征 class PublicOpinionAnalyzer: """ 公众情绪分析器 分析公众讨论的情感倾向和主要话题 """ # 正负面情感词库（示例） POSITIVE_WORDS = [ "支持", "赞同", "优秀", "伟大", "喜欢", "爱", "好", "棒", "support", "great", "love", "good", "excellent", "awesome" ] NEGATIVE_WORDS = [ "反对", "批评", "糟糕", "讨厌", "恨", "坏", "差", "愚蠢", "oppose", "criticize", "bad", "terrible", "hate", "stupid" ] STRONG_EMOTION_WORDS = [ "强烈", "非常", "极其", "绝对", "坚决", "一定", "务必", "strongly", "very", "extremely", "absolutely", "definitely" ] def __init__(self): """初始化公众情绪分析器""" self.social_posts = [] self.comments = [] def load_social_data(self, posts: List[Dict], comments: List[Dict] = None): """ 加载社交媒体数据 Args: posts: 帖子列表 comments: 评论列表（可选） """ self.social_posts = posts self.comments = comments or [] def analyze_sentiment(self, text: str) -> Tuple[str, float, bool]: """ 分析文本情感 Args: text: 文本内容 Returns: (情感倾向, 强度分数, 是否强烈情绪) """ text_lower = text.lower() # 计算正负面词频 positive_count = sum(1 for word in self.POSITIVE_WORDS if word in text_lower) negative_count = sum(1 for word in self.NEGATIVE_WORDS if word in text_lower) # 计算强烈情绪词频 strong_count = sum(1 for word in self.STRONG_EMOTION_WORDS if word in text_lower) is_strong = strong_count > 0 # 计算得分 total = positive_count + negative_count if total == 0: return "neutral", 0.0, is_strong score = (positive_count - negative_count) / total # 归一化强度到[0, 1] intensity = min(1.0, total / 10.0) # 判断情感倾向 if score > 0.2: sentiment = "positive" elif score < -0.2: sentiment = "negative" else: sentiment = "neutral" return sentiment, intensity, is_strong def analyze_overall_sentiment_distribution(self) -> EmotionDistribution: """ 分析整体情感分布 Returns: EmotionDistribution对象 """ all_texts = self.social_posts + self.comments if not all_texts: return EmotionDistribution(0.0, 0.0, 0.0, 0.0) sentiment_counts = Counter() strong_emotion_count = 0 for item in all_texts: text = item.get("content", "") sentiment, intensity, is_strong = self.analyze_sentiment(text) sentiment_counts[sentiment] += 1 if is_strong: strong_emotion_count += 1 total = sum(sentiment_counts.values()) if total == 0: return EmotionDistribution(0.0, 0.0, 0.0, 0.0) return EmotionDistribution( positive=sentiment_counts["positive"] / total, neutral=sentiment_counts["neutral"] / total, negative=sentiment_counts["negative"] / total, strong_emotion_ratio=strong_emotion_count / total ) def extract_topics(self, min_mentions: int = 5) -> List[str]: """ 提取主要话题 Args: min_mentions: 最小提及次数 Returns: 话题列表 """ all_texts = [p.get("content", "") for p in self.social_posts] combined_text = " ".join(all_texts) # 简单的关键词提取（实际应该使用NLP方法） words = re.findall(r'\b[\w\u4e00-\u9fff]{2,}\b', combined_text) word_counts = Counter(words) # 过滤停用词 stop_words = { "的", "了", "在", "是", "和", "有", "说", "这", "那", "the", "and", "of", "to", "in", "is", "it", "that" } for stop_word in stop_words: word_counts.pop(stop_word, None) # 返回高频词作为话题 return [word for word, count in word_counts.most_common(20) if count >= min_mentions] def analyze_topic_sentiment(self, topic: str) -> TopicAnalysis: """ 分析特定话题的情感 Args: topic: 话题关键词 Returns: TopicAnalysis对象 """ topic_posts = [ p for p in self.social_posts if topic.lower() in p.get("content", "").lower() ] if not topic_posts: return TopicAnalysis(topic, 0.0, "neutral", "弱") # 计算提及频率 mention_frequency = len(topic_posts) / len(self.social_posts) if self.social_posts else 0.0 # 分析该话题的整体情感 sentiments = [] intensities = [] for post in topic_posts: sentiment, intensity, _ = self.analyze_sentiment(post.get("content", "")) sentiments.append(sentiment) intensities.append(intensity) # 统计情感倾向 sentiment_counts = Counter(sentiments) dominant_sentiment = sentiment_counts.most_common(1)[0][0] if sentiment_counts else "neutral" # 计算平均强度 avg_intensity = sum(intensities) / len(intensities) if intensities else 0.0 intensity_level = "强" if avg_intensity > 0.6 else ("中" if avg_intensity > 0.3 else "弱") return TopicAnalysis( topic=topic, mention_frequency=mention_frequency, sentiment=dominant_sentiment, intensity=intensity_level ) def estimate_support_rate(self) -> Dict[str, float]: """ 估算公众支持率 Returns: 支持率字典 """ sentiment_dist = self.analyze_overall_sentiment_distribution() # 基于情感分布估算支持率（简化版） # 实际应该结合社交媒体粉丝数、互动量等数据 support_rate = sentiment_dist.positive opposition_rate = sentiment_dist.negative neutral_rate = sentiment_dist.neutral return { "整体支持率": support_rate, "反对比例": opposition_rate, "中立比例": neutral_rate, "极化指数": abs(support_rate - opposition_rate) } def generate_public_opinion_report(self) -> Dict: """ 生成公众情绪分析报告 Returns: 分析报告字典 """ # 分析情感分布 sentiment_dist = self.analyze_overall_sentiment_distribution() # 提取主要话题 topics = self.extract_topics() # 分析每个话题的情感 topic_analyses = [self.analyze_topic_sentiment(topic) for topic in topics[:10]] # 估算支持率 support_rates = self.estimate_support_rate() report = { "情感分布": { "正面": f"{sentiment_dist.positive:.1%}", "中立": f"{sentiment_dist.neutral:.1%}", "负面": f"{sentiment_dist.negative:.1%}", "强烈情绪占比": f"{sentiment_dist.strong_emotion_ratio:.1%}" }, "主要话题": [ { "话题": ta.topic, "提及频率": f"{ta.mention_frequency:.1%}", "情感倾向": ta.sentiment, "情感强度": ta.intensity } for ta in topic_analyses ], "支持率估算": { "整体支持率": f"{support_rates['整体支持率']:.1%}", "反对比例": f"{support_rates['反对比例']:.1%}", "中立比例": f"{support_rates['中立比例']:.1%}", "极化指数": f"{support_rates['极化指数']:.2f}" } } return report def main(): """测试公众情绪分析器""" analyzer = PublicOpinionAnalyzer() # 示例社交媒体数据 posts = [ { "content": "我强烈支持他的政策，他会让国家变得更伟大！", "date": "2024-01-01" }, { "content": "他的政策简直是一场灾难，完全错误。", "date": "2024-01-02" }, { "content": "我不确定他的政策效果如何，需要再观察。", "date": "2024-01-03" }, { "content": "非常好！支持！", "date": "2024-01-04" }, { "content": "太糟糕了，反对到底！", "date": "2024-01-05" } ] analyzer.load_social_data(posts) # 生成报告 print("=== 公众情绪分析报告 ===") report = analyzer.generate_public_opinion_report() print(json.dumps(report, ensure_ascii=False, indent=2)) if __name__ == "__main__": main()