diff --git a/README-EN.md b/README-EN.md
new file mode 100644
index 0000000..e925c32
--- /dev/null
+++ b/README-EN.md
@@ -0,0 +1,655 @@
+
+
+# 📊 Weibo Public Opinion Multi-Agent Analysis System
+
+
+
+[](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/stargazers)
+[](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/network)
+[](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/issues)
+[](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/blob/main/LICENSE)
+
+[English](./README-EN.md) | [中文文档](./README.md)
+
+
+
+
Flask + Streamlit]
+ end
+
+ subgraph "Multi-Agent Collaboration Layer"
+ QE[QueryEngine
News Search Agent]
+ ME[MediaEngine
Multimedia Search Agent]
+ IE[InsightEngine
Deep Insight Agent]
+ RE[ReportEngine
Report Generation Agent]
+ Forum[ForumEngine
Agent Forum Communication Center]
+ end
+
+ subgraph "Data Processing Layer"
+ MS[MindSpider
Weibo Crawler System]
+ SA[SentimentAnalysis
Sentiment Analysis Model Collection]
+ DB[(MySQL
Database)]
+ end
+
+ subgraph "External Service Layer"
+ LLM[LLM API
DeepSeek/Kimi/Gemini]
+ Search[Search API
Tavily/Bocha]
+ end
+
+ UI --> QE
+ UI --> ME
+ UI --> IE
+ UI --> RE
+
+ QE --> Search
+ ME --> Search
+ IE --> MS
+ IE --> SA
+
+ QE --> LLM
+ ME --> LLM
+ IE --> LLM
+ RE --> LLM
+
+ MS --> DB
+ SA --> DB
+
+ %% Agent Forum Communication Mechanism
+ QE <--> Forum
+ ME <--> Forum
+ IE <--> Forum
+ RE <--> Forum
+```
+
+### Agent Collaboration Workflow
+
+The system's core workflow is based on multi-agent collaboration:
+
+1. **QueryEngine (News Query Agent)**: Uses Tavily API to search authoritative news reports, providing official information sources
+2. **MediaEngine (Multimedia Search Agent)**: Conducts multimodal content search through Bocha API to gather social media perspectives
+3. **InsightEngine (Deep Insight Agent)**: Queries local Weibo database, combines multiple sentiment analysis models for deep analysis
+4. **ForumEngine (Forum Monitoring Agent)**: Real-time monitoring of agent log outputs, extracts key information and promotes collaboration
+5. **ReportEngine (Report Generation Agent)**: Based on analysis results from all agents, uses Gemini LLM to generate comprehensive HTML reports
+
+### Project Code Structure
+
+```
+Weibo_PublicOpinion_AnalysisSystem/
+├── QueryEngine/ # News Query Engine Agent
+│ ├── agent.py # Agent main logic
+│ ├── llms/ # LLM interface wrapper
+│ ├── nodes/ # Processing nodes
+│ ├── tools/ # Search tools
+│ └── utils/ # Utility functions
+├── MediaEngine/ # Multimedia Search Engine Agent
+│ ├── agent.py # Agent main logic
+│ ├── llms/ # LLM interfaces
+│ ├── tools/ # Search tools
+│ └── ... # Other modules
+├── InsightEngine/ # Data Insight Engine Agent
+│ ├── agent.py # Agent main logic
+│ ├── llms/ # LLM interface wrapper
+│ │ ├── deepseek.py # DeepSeek API
+│ │ ├── kimi.py # Kimi API
+│ │ ├── openai_llm.py # OpenAI format API
+│ │ └── base.py # LLM base class
+│ ├── nodes/ # Processing nodes
+│ │ ├── first_search_node.py # First search node
+│ │ ├── reflection_node.py # Reflection node
+│ │ ├── summary_nodes.py # Summary nodes
+│ │ ├── search_node.py # Search node
+│ │ ├── sentiment_node.py # Sentiment analysis node
+│ │ └── insight_node.py # Insight generation node
+│ ├── tools/ # Database query and analysis tools
+│ │ ├── media_crawler_db.py # Database query tool
+│ │ └── sentiment_analyzer.py # Sentiment analysis integration tool
+│ ├── state/ # State management
+│ │ ├── __init__.py
+│ │ └── state.py # Agent state definition
+│ ├── prompts/ # Prompt templates
+│ │ ├── __init__.py
+│ │ └── prompts.py # Various prompts
+│ └── utils/ # Utility functions
+│ ├── __init__.py
+│ ├── config.py # Configuration management
+│ └── helpers.py # Helper functions
+├── ReportEngine/ # Report Generation Engine Agent
+│ ├── agent.py # Agent main logic
+│ ├── llms/ # LLM interfaces
+│ │ └── gemini.py # Gemini API dedicated
+│ ├── nodes/ # Report generation nodes
+│ │ ├── template_selection.py # Template selection node
+│ │ └── html_generation.py # HTML generation node
+│ ├── report_template/ # Report template library
+│ │ ├── 社会公共热点事件分析.md
+│ │ ├── 商业品牌舆情监测.md
+│ │ └── ... # More templates
+│ └── flask_interface.py # Flask API interface
+├── ForumEngine/ # Forum Communication Engine Agent
+│ └── monitor.py # Log monitoring and forum management
+├── MindSpider/ # Weibo Crawler System
+│ ├── main.py # Crawler main program
+│ ├── BroadTopicExtraction/ # Topic extraction module
+│ │ ├── get_today_news.py # Today's news fetching
+│ │ └── topic_extractor.py # Topic extractor
+│ ├── DeepSentimentCrawling/ # Deep sentiment crawling
+│ │ ├── MediaCrawler/ # Media crawler core
+│ │ └── platform_crawler.py # Platform crawler management
+│ └── schema/ # Database schema
+│ └── init_database.py # Database initialization
+├── SentimentAnalysisModel/ # Sentiment Analysis Model Collection
+│ ├── WeiboSentiment_Finetuned/ # Fine-tuned BERT/GPT-2 models
+│ ├── WeiboMultilingualSentiment/ # Multilingual sentiment analysis
+│ ├── WeiboSentiment_SmallQwen/ # Small Qwen model
+│ └── WeiboSentiment_MachineLearning/ # Traditional machine learning methods
+├── SingleEngineApp/ # Individual Agent Streamlit apps
+│ ├── query_engine_streamlit_app.py
+│ ├── media_engine_streamlit_app.py
+│ └── insight_engine_streamlit_app.py
+├── templates/ # Flask templates
+│ └── index.html # Main interface template
+├── static/ # Static resources
+├── logs/ # Runtime log directory
+├── app.py # Flask main application entry
+├── config.py # Global configuration file
+└── requirements.txt # Python dependency list
+```
+
+## 🚀 Quick Start
+
+### System Requirements
+
+- **Operating System**: Windows 10/11 (Linux/macOS also supported)
+- **Python Version**: 3.11+
+- **Conda**: Anaconda or Miniconda
+- **Database**: MySQL 8.0+ (or choose our cloud database service)
+- **Memory**: 8GB+ recommended
+
+### 1. Create Conda Environment
+
+```bash
+# Create conda environment named pytorch_python11
+conda create -n pytorch_python11 python=3.11
+conda activate pytorch_python11
+```
+
+### 2. Install Dependencies
+
+```bash
+# Install basic dependencies
+pip install -r requirements.txt
+
+# If you need local sentiment analysis functionality, install PyTorch
+# CPU version
+pip install torch torchvision torchaudio
+
+# CUDA 11.8 version (if you have GPU)
+pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
+
+# Install transformers and other AI-related dependencies
+pip install transformers scikit-learn xgboost
+```
+
+### 3. Install Playwright Browser Drivers
+
+```bash
+# Install browser drivers (for crawler functionality)
+playwright install chromium
+```
+
+### 4. System Configuration
+
+#### 4.1 Configure API Keys
+
+Edit the `config.py` file and fill in your API keys:
+
+```python
+# MySQL Database Configuration
+DB_HOST = "localhost"
+DB_PORT = 3306
+DB_USER = "your_username"
+DB_PASSWORD = "your_password"
+DB_NAME = "weibo_analysis"
+DB_CHARSET = "utf8mb4"
+
+# DeepSeek API (Apply at: https://www.deepseek.com/)
+DEEPSEEK_API_KEY = "your_deepseek_api_key"
+
+# Tavily Search API (Apply at: https://www.tavily.com/)
+TAVILY_API_KEY = "your_tavily_api_key"
+
+# Kimi API (Apply at: https://www.kimi.com/)
+KIMI_API_KEY = "your_kimi_api_key"
+
+# Gemini API (Apply at: https://api.chataiapi.com/)
+GEMINI_API_KEY = "your_gemini_api_key"
+
+# Bocha Search API (Apply at: https://open.bochaai.com/)
+BOCHA_Web_Search_API_KEY = "your_bocha_api_key"
+
+# Silicon Flow API (Apply at: https://siliconflow.cn/)
+GUIJI_QWEN3_API_KEY = "your_guiji_api_key"
+```
+
+#### 4.2 Database Initialization
+
+**Option 1: Use Local Database**
+```bash
+# Local MySQL database initialization
+cd MindSpider
+python schema/init_database.py
+```
+
+**Option 2: Use Cloud Database Service (Recommended)**
+
+We provide convenient cloud database service with 100,000+ daily real Weibo data, currently **free application** during the promotion period!
+
+- Real Weibo data, updated in real-time
+- Pre-processed sentiment annotation data
+- Multi-dimensional tag classification
+- High-availability cloud service
+- Professional technical support
+
+**Contact us to apply for free cloud database access: 📧 670939375@qq.com**
+
+### 5. Launch System
+
+#### 5.1 Complete System Launch (Recommended)
+
+```bash
+# In project root directory, activate conda environment
+conda activate pytorch_python11
+
+# Start main application (automatically starts all agents)
+python app.py
+```
+
+Visit http://localhost:5000 to use the complete system
+
+#### 5.2 Launch Individual Agents
+
+```bash
+# Start QueryEngine
+streamlit run SingleEngineApp/query_engine_streamlit_app.py --server.port 8503
+
+# Start MediaEngine
+streamlit run SingleEngineApp/media_engine_streamlit_app.py --server.port 8502
+
+# Start InsightEngine
+streamlit run SingleEngineApp/insight_engine_streamlit_app.py --server.port 8501
+```
+
+#### 5.3 Standalone Crawler System
+
+```bash
+# Enter crawler directory
+cd MindSpider
+
+# Project initialization
+python main.py --setup
+
+# Run complete crawler workflow
+python main.py --complete --date 2024-01-20
+
+# Run topic extraction only
+python main.py --broad-topic --date 2024-01-20
+
+# Run deep crawling only
+python main.py --deep-sentiment --platforms xhs dy wb
+```
+
+## 💾 Database Configuration
+
+### Local Database Configuration
+
+1. **Install MySQL 8.0+**
+2. **Create Database**:
+ ```sql
+ CREATE DATABASE weibo_analysis CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci;
+ ```
+3. **Run Initialization Script**:
+ ```bash
+ cd MindSpider
+ python schema/init_database.py
+ ```
+
+### Auto-Crawling Configuration
+
+Configure automatic crawling tasks for continuous data updates:
+
+```python
+# Configure crawler parameters in MindSpider/config.py
+CRAWLER_CONFIG = {
+ 'max_pages': 200, # Maximum pages to crawl
+ 'delay': 1, # Request delay (seconds)
+ 'timeout': 30, # Timeout (seconds)
+ 'platforms': ['xhs', 'dy', 'wb', 'bili'], # Crawling platforms
+ 'daily_keywords': 100, # Daily keywords count
+ 'max_notes_per_keyword': 50, # Max content per keyword
+ 'use_proxy': False, # Whether to use proxy
+}
+```
+
+### Cloud Database Service (Recommended)
+
+**Why Choose Our Cloud Database Service?**
+
+- **Rich Data Sources**: 100,000+ daily real Weibo data covering hot topics across all industries
+- **High-Quality Annotations**: Professional team manually annotated sentiment data with 95%+ accuracy
+- **Multi-Dimensional Analysis**: Including topic classification, sentiment tendency, influence scoring and other multi-dimensional tags
+- **Real-Time Updates**: 24/7 continuous data collection ensuring timeliness
+- **Technical Support**: Professional team providing technical support and customization services
+
+**Application Method**:
+📧 Email Contact: 670939375@qq.com
+📝 Email Subject: Apply for Weibo Public Opinion Cloud Database Access
+📝 Email Content: Please describe your use case and expected data volume requirements
+
+**Promotion Period Benefits**:
+- Free basic cloud database access
+- Free technical support and deployment guidance
+- Priority access to new features
+
+## ⚙️ Advanced Configuration
+
+### Modify Key Parameters
+
+#### Agent Configuration Parameters
+
+Each agent has dedicated configuration files that can be adjusted according to needs:
+
+```python
+# QueryEngine/utils/config.py
+class Config:
+ max_reflections = 2 # Reflection rounds
+ max_search_results = 15 # Maximum search results
+ max_content_length = 8000 # Maximum content length
+
+# MediaEngine/utils/config.py
+class Config:
+ comprehensive_search_limit = 10 # Comprehensive search limit
+ web_search_limit = 15 # Web search limit
+
+# InsightEngine/utils/config.py
+class Config:
+ default_search_topic_globally_limit = 200 # Global search limit
+ default_get_comments_limit = 500 # Comment retrieval limit
+ max_search_results_for_llm = 50 # Max results for LLM
+```
+
+#### Sentiment Analysis Model Configuration
+
+```python
+# InsightEngine/tools/sentiment_analyzer.py
+SENTIMENT_CONFIG = {
+ 'model_type': 'multilingual', # Options: 'bert', 'multilingual', 'qwen'
+ 'confidence_threshold': 0.8, # Confidence threshold
+ 'batch_size': 32, # Batch size
+ 'max_sequence_length': 512, # Max sequence length
+}
+```
+
+### Integrate Different LLM Models
+
+The system supports multiple LLM providers, switchable in each agent's configuration:
+
+```python
+# Configure in each Engine's utils/config.py
+class Config:
+ default_llm_provider = "deepseek" # Options: "deepseek", "openai", "kimi", "gemini"
+
+ # DeepSeek configuration
+ deepseek_api_key = "your_api_key"
+ deepseek_model = "deepseek-chat"
+
+ # OpenAI compatible configuration
+ openai_api_key = "your_api_key"
+ openai_model = "gpt-3.5-turbo"
+ openai_base_url = "https://api.openai.com/v1"
+
+ # Kimi configuration
+ kimi_api_key = "your_api_key"
+ kimi_model = "moonshot-v1-8k"
+
+ # Gemini configuration
+ gemini_api_key = "your_api_key"
+ gemini_model = "gemini-pro"
+```
+
+### Change Sentiment Analysis Models
+
+The system integrates multiple sentiment analysis methods, selectable based on needs:
+
+#### 1. BERT-based Fine-tuned Model (Highest Accuracy)
+
+```bash
+# Use BERT Chinese model
+cd SentimentAnalysisModel/WeiboSentiment_Finetuned/BertChinese-Lora
+python predict.py --text "This product is really great"
+```
+
+#### 2. GPT-2 LoRA Fine-tuned Model (Faster Speed)
+
+```bash
+cd SentimentAnalysisModel/WeiboSentiment_Finetuned/GPT2-Lora
+python predict.py --text "I'm not feeling great today"
+```
+
+#### 3. Small Qwen Model (Balanced)
+
+```bash
+cd SentimentAnalysisModel/WeiboSentiment_SmallQwen
+python predict_universal.py --text "This event was very successful"
+```
+
+#### 4. Traditional Machine Learning Methods (Lightweight)
+
+```bash
+cd SentimentAnalysisModel/WeiboSentiment_MachineLearning
+python predict.py --model_type "svm" --text "Service attitude needs improvement"
+```
+
+#### 5. Multilingual Sentiment Analysis (Supports 22 Languages)
+
+```bash
+cd SentimentAnalysisModel/WeiboMultilingualSentiment
+python predict.py --text "This product is amazing!" --lang "en"
+```
+
+### Integrate Custom Business Database
+
+#### 1. Modify Database Connection Configuration
+
+```python
+# Add your business database configuration in config.py
+BUSINESS_DB_HOST = "your_business_db_host"
+BUSINESS_DB_PORT = 3306
+BUSINESS_DB_USER = "your_business_user"
+BUSINESS_DB_PASSWORD = "your_business_password"
+BUSINESS_DB_NAME = "your_business_database"
+```
+
+#### 2. Create Custom Data Access Tools
+
+```python
+# InsightEngine/tools/custom_db_tool.py
+class CustomBusinessDBTool:
+ """Custom business database query tool"""
+
+ def __init__(self):
+ self.connection_config = {
+ 'host': config.BUSINESS_DB_HOST,
+ 'port': config.BUSINESS_DB_PORT,
+ 'user': config.BUSINESS_DB_USER,
+ 'password': config.BUSINESS_DB_PASSWORD,
+ 'database': config.BUSINESS_DB_NAME,
+ }
+
+ def search_business_data(self, query: str, table: str):
+ """Query business data"""
+ # Implement your business logic
+ pass
+
+ def get_customer_feedback(self, product_id: str):
+ """Get customer feedback data"""
+ # Implement customer feedback query logic
+ pass
+```
+
+#### 3. Integrate into InsightEngine
+
+```python
+# Integrate custom tools in InsightEngine/agent.py
+from .tools.custom_db_tool import CustomBusinessDBTool
+
+class DeepSearchAgent:
+ def __init__(self, config=None):
+ # ... other initialization code
+ self.custom_db_tool = CustomBusinessDBTool()
+
+ def execute_custom_search(self, query: str):
+ """Execute custom business data search"""
+ return self.custom_db_tool.search_business_data(query, "your_table")
+```
+
+### Custom Report Templates
+
+#### 1. Create Template Files
+
+Create new Markdown templates in the `ReportEngine/report_template/` directory:
+
+```markdown
+
+# Enterprise Brand Public Opinion Monitoring Report
+
+## 📊 Executive Summary
+{executive_summary}
+
+## 🔍 Brand Mention Analysis
+### Mention Volume Trends
+{mention_trend}
+
+### Sentiment Distribution
+{sentiment_distribution}
+
+## 📈 Competitor Analysis
+{competitor_analysis}
+
+## 🎯 Key Insights Summary
+{key_insights}
+
+## ⚠️ Risk Alerts
+{risk_alerts}
+
+## 📋 Improvement Recommendations
+{recommendations}
+
+---
+*Report Type: Enterprise Brand Public Opinion Monitoring*
+*Generation Time: {generation_time}*
+*Data Sources: {data_sources}*
+```
+
+#### 2. Use in Web Interface
+
+The system supports uploading custom template files (.md or .txt format), selectable when generating reports.
+
+## 🤝 Contributing Guide
+
+We welcome all forms of contributions!
+
+### How to Contribute
+
+1. **Fork the project** to your GitHub account
+2. **Create Feature branch**: `git checkout -b feature/AmazingFeature`
+3. **Commit changes**: `git commit -m 'Add some AmazingFeature'`
+4. **Push to branch**: `git push origin feature/AmazingFeature`
+5. **Open Pull Request**
+
+### Contribution Types
+
+- 🐛 Bug fixes
+- ✨ New feature development
+- 📚 Documentation improvements
+- 🎨 UI/UX improvements
+- ⚡ Performance optimization
+- 🧪 Test case additions
+
+### Development Standards
+
+- Code follows PEP8 standards
+- Commit messages use clear Chinese/English descriptions
+- New features need corresponding test cases
+- Update related documentation
+
+## 📄 License
+
+This project is licensed under the [MIT License](LICENSE). Please see the LICENSE file for details.
+
+## 🎉 Support & Contact
+
+### Get Help
+
+- **Project Homepage**: [GitHub Repository](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem)
+- **Issue Reporting**: [Issues Page](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/issues)
+- **Feature Requests**: [Discussions Page](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/discussions)
+
+### Contact Information
+
+- 📧 **Email**: 670939375@qq.com
+- 💬 **QQ Group**: [Join Technical Discussion Group]
+- 🐦 **WeChat**: [Scan QR Code for Technical Support]
+
+### Business Cooperation
+
+- 🏢 **Enterprise Custom Development**
+- 📊 **Big Data Services**
+- 🎓 **Academic Collaboration**
+- 💼 **Technical Training**
+
+### Cloud Service Application
+
+**Free Cloud Database Service Application**:
+📧 Send email to: 670939375@qq.com
+📝 Subject: Weibo Public Opinion Cloud Database Application
+📝 Description: Your use case and requirements
+
+## 👥 Contributors
+
+Thanks to these excellent contributors:
+
+[](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/graphs/contributors)
+
+---
+
+
+
+**⭐ If this project helps you, please give us a star!**
+
+Made with ❤️ by [Weibo Public Opinion Analysis Team](https://github.com/666ghj)
+
+
\ No newline at end of file
diff --git a/README.md b/README.md
index ce45d26..983125e 100644
--- a/README.md
+++ b/README.md
@@ -1,35 +1,37 @@
-
+
-
+# 微舆 - 致力于打造简洁通用的舆情分析平台
- [](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/stargazers)
- [](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/network)
- [](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/issues)
- [](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/graphs/contributors)
- [](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/blob/main/LICENSE)
+[](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/stargazers)
+[](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/network)
+[](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/issues)
+[](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/blob/main/LICENSE)
+
+[English](./README-EN.md) | [中文文档](./README.md)
-
+
微博爬虫系统]
- SA[SentimentAnalysis
情感分析模型]
+ SA[SentimentAnalysis
情感分析模型集合]
DB[(MySQL
数据库)]
end
@@ -81,129 +83,110 @@ graph TB
ME <--> Forum
IE <--> Forum
RE <--> Forum
-
- style UI fill:#e1f5fe
- style QE fill:#fff3e0
- style ME fill:#fff3e0
- style IE fill:#fff3e0
- style RE fill:#f3e5f5
- style Forum fill:#e8f5e9
- style MS fill:#fce4ec
- style SA fill:#fce4ec
- style DB fill:#fff9c4
- style LLM fill:#e3f2fd
- style Search fill:#e3f2fd
```
-### 数据流程图
+### Agent协作流程
-```mermaid
-sequenceDiagram
- participant User as 用户
- participant UI as Web界面
- participant QE as QueryEngine
- participant ME as MediaEngine
- participant IE as InsightEngine
- participant Forum as ForumEngine
- participant RE as ReportEngine
- participant DB as 数据库
-
- User->>UI: 输入查询关键词
- UI->>QE: 发起搜索请求
- UI->>ME: 发起搜索请求
- UI->>IE: 发起搜索请求
-
- Note over QE,IE: Agent执行前先读取论坛信息
- QE->>Forum: 读取论坛交流信息
- ME->>Forum: 读取论坛交流信息
- IE->>Forum: 读取论坛交流信息
-
- par 并行处理与持续思维链交流
- Note over QE: 结构思考→反思搜索→持续交流
- QE->>QE: 确定新闻搜索结构
- QE->>Forum: 思维链交流(结构思考)
- QE->>QE: 多步反思与搜索分析
- QE->>Forum: 思维链交流(搜索进展)
- QE->>QE: 生成汇总报告
- QE->>Forum: 思维链交流(关键发现)
- and
- Note over ME: 结构思考→反思搜索→持续交流
- ME->>ME: 确定多媒体搜索结构
- ME->>Forum: 思维链交流(结构思考)
- ME->>ME: 多步反思与搜索分析
- ME->>Forum: 思维链交流(搜索进展)
- ME->>ME: 生成汇总报告
- ME->>Forum: 思维链交流(关键发现)
- and
- Note over IE: 结构思考→反思搜索→持续交流
- IE->>IE: 确定洞察分析结构
- IE->>Forum: 思维链交流(结构思考)
- IE->>DB: 查询微博数据
- IE->>IE: 多步反思与情感洞察
- IE->>Forum: 思维链交流(洞察进展)
- IE->>IE: 生成汇总报告
- IE->>Forum: 思维链交流(关键发现)
- end
-
- Note over Forum: 论坛汇总Agent交流信息
- Forum->>RE: 触发报告生成
- RE->>Forum: 读取所有Agent的交流信息
- RE->>QE: 获取QueryEngine汇总报告
- RE->>ME: 获取MediaEngine汇总报告
- RE->>IE: 获取InsightEngine汇总报告
-
- Note over RE: ReportEngine智能报告生成
- RE->>RE: 读取模板库与样式库并选择
- RE->>RE: 分步思考生成报告各部分
- RE->>RE: 整合生成最终报告
- RE->>UI: 生成综合HTML报告
- UI->>User: 展示分析结果
-```
+系统核心工作流程基于多Agent协作模式：
-## 项目结构
+1. **QueryEngine（新闻查询Agent）**：使用Tavily API搜索权威新闻报道，提供官方信息源
+2. **MediaEngine（多媒体搜索Agent）**：通过Bocha API进行多模态内容搜索，获取社交媒体观点
+3. **InsightEngine（深度洞察Agent）**：查询本地微博数据库，结合多种情感分析模型进行深度分析
+4. **ForumEngine（论坛监控Agent）**：实时监控各Agent日志输出，提取关键信息并促进协作
+5. **ReportEngine（报告生成Agent）**：基于所有Agent的分析结果，使用Gemini LLM生成综合HTML报告
+
+### 项目代码结构
```
Weibo_PublicOpinion_AnalysisSystem/
-├── QueryEngine/ # web查询引擎Agent
-│ ├── agent.py # Agent主逻辑
-│ ├── llms/ # LLM接口封装
-│ ├── nodes/ # 处理节点
-│ ├── tools/ # 搜索工具
-│ └── utils/ # 工具函数
-├── MediaEngine/ # 媒体引擎Agent
-│ └── (类似结构)
-├── InsightEngine/ # 数据库引擎Agent
-│ └── (类似结构)
-├── ReportEngine/ # 报告生成Agent
-│ ├── report_template/ # 报告模板
-│ └── flask_interface.py # API接口
-├── ForumEgine/ # 论坛交流Agent
-│ └── monitor.py # 论坛交流管理器
-├── MindSpider/ # 微博爬虫系统
-│ ├── BroadTopicExtraction/ # 话题提取
-│ ├── DeepSentimentCrawling/ # 深度爬取
-│ └── schema/ # 数据库结构
-├── SentimentAnalysisModel/ # 情感分析模型
-│ ├── BertTopicDetection_Finetuned/
-│ ├── WeiboSentiment_Finetuned/
-│ └── WeiboSentiment_MachineLearning/
-├── SingleEngineApp/ # Streamlit应用
-├── templates/ # Flask模板
-├── static/ # 静态资源
-├── logs/ # 运行日志
-├── app.py # 主应用入口
-├── config.py # 配置文件
-└── requirements.txt # 依赖包
+├── QueryEngine/ # 新闻查询引擎Agent
+│ ├── agent.py # Agent主逻辑
+│ ├── llms/ # LLM接口封装
+│ ├── nodes/ # 处理节点
+│ ├── tools/ # 搜索工具
+│ └── utils/ # 工具函数
+├── MediaEngine/ # 多媒体搜索引擎Agent
+│ ├── agent.py # Agent主逻辑
+│ ├── llms/ # LLM接口
+│ ├── tools/ # 搜索工具
+│ └── ... # 其他模块
+├── InsightEngine/ # 数据洞察引擎Agent
+│ ├── agent.py # Agent主逻辑
+│ ├── llms/ # LLM接口封装
+│ │ ├── deepseek.py # DeepSeek API
+│ │ ├── kimi.py # Kimi API
+│ │ ├── openai_llm.py # OpenAI格式API
+│ │ └── base.py # LLM基类
+│ ├── nodes/ # 处理节点
+│ │ ├── first_search_node.py # 首次搜索节点
+│ │ ├── reflection_node.py # 反思节点
+│ │ ├── summary_nodes.py # 总结节点
+│ │ ├── search_node.py # 搜索节点
+│ │ ├── sentiment_node.py # 情感分析节点
+│ │ └── insight_node.py # 洞察生成节点
+│ ├── tools/ # 数据库查询和分析工具
+│ │ ├── media_crawler_db.py # 数据库查询工具
+│ │ └── sentiment_analyzer.py # 情感分析集成工具
+│ ├── state/ # 状态管理
+│ │ ├── __init__.py
+│ │ └── state.py # Agent状态定义
+│ ├── prompts/ # 提示词模板
+│ │ ├── __init__.py
+│ │ └── prompts.py # 各类提示词
+│ └── utils/ # 工具函数
+│ ├── __init__.py
+│ ├── config.py # 配置管理
+│ └── helpers.py # 辅助函数
+├── ReportEngine/ # 报告生成引擎Agent
+│ ├── agent.py # Agent主逻辑
+│ ├── llms/ # LLM接口
+│ │ └── gemini.py # Gemini API专用
+│ ├── nodes/ # 报告生成节点
+│ │ ├── template_selection.py # 模板选择节点
+│ │ └── html_generation.py # HTML生成节点
+│ ├── report_template/ # 报告模板库
+│ │ ├── 社会公共热点事件分析.md
+│ │ ├── 商业品牌舆情监测.md
+│ │ └── ... # 更多模板
+│ └── flask_interface.py # Flask API接口
+├── ForumEngine/ # 论坛交流引擎Agent
+│ └── monitor.py # 日志监控和论坛管理
+├── MindSpider/ # 微博爬虫系统
+│ ├── main.py # 爬虫主程序
+│ ├── BroadTopicExtraction/ # 话题提取模块
+│ │ ├── get_today_news.py # 今日新闻获取
+│ │ └── topic_extractor.py # 话题提取器
+│ ├── DeepSentimentCrawling/ # 深度情感爬取
+│ │ ├── MediaCrawler/ # 媒体爬虫核心
+│ │ └── platform_crawler.py # 平台爬虫管理
+│ └── schema/ # 数据库结构
+│ └── init_database.py # 数据库初始化
+├── SentimentAnalysisModel/ # 情感分析模型集合
+│ ├── WeiboSentiment_Finetuned/ # 微调BERT/GPT-2模型
+│ ├── WeiboMultilingualSentiment/ # 多语言情感分析
+│ ├── WeiboSentiment_SmallQwen/ # 小型Qwen模型
+│ └── WeiboSentiment_MachineLearning/ # 传统机器学习方法
+├── SingleEngineApp/ # 单独Agent的Streamlit应用
+│ ├── query_engine_streamlit_app.py
+│ ├── media_engine_streamlit_app.py
+│ └── insight_engine_streamlit_app.py
+├── templates/ # Flask模板
+│ └── index.html # 主界面模板
+├── static/ # 静态资源
+├── logs/ # 运行日志目录
+├── app.py # Flask主应用入口
+├── config.py # 全局配置文件
+└── requirements.txt # Python依赖包清单
```
-## 快速开始
+## 🚀 快速开始
### 环境要求
-- **操作系统**: Windows 10/11
+- **操作系统**: Windows 10/11（Linux/macOS也支持）
- **Python版本**: 3.11+
- **Conda**: Anaconda或Miniconda
-- **数据库**: MySQL 8.0+
+- **数据库**: MySQL 8.0+（可选择我们的云数据库服务）
- **内存**: 建议8GB以上
### 1. 创建Conda环境
@@ -220,14 +203,14 @@ conda activate pytorch_python11
# 基础依赖安装
pip install -r requirements.txt
-# 如果需要情感分析功能，安装PyTorch（根据CUDA版本选择）
+# 如果需要本地情感分析功能，安装PyTorch
# CPU版本
pip install torch torchvision torchaudio
-# CUDA 11.8版本
+# CUDA 11.8版本（如有GPU）
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
-# 安装transformers（用于BERT/GPT模型）
+# 安装transformers等AI相关依赖
pip install transformers scikit-learn xgboost
```
@@ -272,16 +255,30 @@ BOCHA_Web_Search_API_KEY = "your_bocha_api_key"
GUIJI_QWEN3_API_KEY = "your_guiji_api_key"
```
-#### 4.2 初始化数据库
+#### 4.2 数据库初始化
+**选择1：使用本地数据库**
```bash
+# 本地MySQL数据库初始化
cd MindSpider
python schema/init_database.py
```
+**选择2：使用云数据库服务（推荐）**
+
+我们提供便捷的云数据库服务，包含日均10万+真实微博数据，目前推广期间**免费申请**！
+
+- 真实微博数据，实时更新
+- 预处理的情感标注数据
+- 多维度标签分类
+- 高可用云端服务
+- 专业技术支持
+
+**联系我们申请免费云数据库访问：📧 670939375@qq.com**
+
### 5. 启动系统
-#### 方式一：完整系统启动（推荐）
+#### 5.1 完整系统启动（推荐）
```bash
# 在项目根目录下，激活conda环境
@@ -291,9 +288,9 @@ conda activate pytorch_python11
python app.py
```
-访问 http://localhost:5000 即可使用系统
+访问 http://localhost:5000 即可使用完整系统
-#### 方式二：单独启动某个Agent
+#### 5.2 单独启动某个Agent
```bash
# 启动QueryEngine
@@ -306,147 +303,353 @@ streamlit run SingleEngineApp/media_engine_streamlit_app.py --server.port 8502
streamlit run SingleEngineApp/insight_engine_streamlit_app.py --server.port 8501
```
-## 使用指南
+#### 5.3 爬虫系统单独使用
-### 基础使用流程
+```bash
+# 进入爬虫目录
+cd MindSpider
-1. **启动系统**：运行 `python app.py`，系统会自动启动所有Agent
+# 项目初始化
+python main.py --setup
-2. **输入查询**：在Web界面搜索框输入要分析的舆情关键词
+# 运行完整爬虫流程
+python main.py --complete --date 2024-01-20
-3. **Agent协作**：
- - QueryEngine：搜索新闻和官方报道，将关键发现发布到论坛
- - MediaEngine：搜索多媒体内容，与其他Agent分享重要信息
- - InsightEngine：分析微博数据和情感，在论坛中交流洞察
- - ForumEngine：提供Agent间交流平台，汇总协作信息
+# 仅运行话题提取
+python main.py --broad-topic --date 2024-01-20
-4. **查看结果**：
- - Agent论坛交流：查看Agent间的实时信息交换
- - 分析报告：查看基于Agent协作的综合HTML报告
+# 仅运行深度爬取
+python main.py --deep-sentiment --platforms xhs dy wb
+```
-### 高级配置
+## 💾 数据库配置
-#### 配置爬虫系统
+### 本地数据库配置
+
+1. **安装MySQL 8.0+**
+2. **创建数据库**：
+ ```sql
+ CREATE DATABASE weibo_analysis CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci;
+ ```
+3. **运行初始化脚本**：
+ ```bash
+ cd MindSpider
+ python schema/init_database.py
+ ```
+
+### 自动爬取配置
+
+配置自动爬取任务，实现数据的持续更新：
-1. **配置爬虫参数**：
```python
-# MindSpider/config.py
+# MindSpider/config.py 中配置爬虫参数
CRAWLER_CONFIG = {
- 'max_pages': 100, # 最大爬取页数
- 'delay': 1, # 请求延迟（秒）
- 'timeout': 30, # 超时时间（秒）
- 'use_proxy': False, # 是否使用代理
+ 'max_pages': 200, # 最大爬取页数
+ 'delay': 1, # 请求延迟（秒）
+ 'timeout': 30, # 超时时间（秒）
+ 'platforms': ['xhs', 'dy', 'wb', 'bili'], # 爬取平台
+ 'daily_keywords': 100, # 每日关键词数量
+ 'max_notes_per_keyword': 50, # 每关键词最大内容数
+ 'use_proxy': False, # 是否使用代理
}
```
-2. **运行爬虫**：
-```bash
-cd MindSpider
-python main.py --topic "话题关键词" --days 7
+### 云数据库服务（推荐）
+
+**为什么选择我们的云数据库服务？**
+
+- **丰富数据源**：日均10万+真实微博数据，涵盖各行业热点话题
+- **高质量标注**：专业团队人工标注的情感数据，准确率95%+
+- **多维度分析**：包含话题分类、情感倾向、影响力评分等多维标签
+- **实时更新**：24小时不间断数据采集，确保时效性
+- **技术支持**：专业团队提供技术支持和定制化服务
+
+**申请方式**：
+📧 邮件联系：670939375@qq.com
+📝 邮件标题：申请微博舆情云数据库访问
+📝 邮件内容：请说明您的使用场景和预期数据量需求
+
+**推广期福利**：
+- 免费提供基础版云数据库访问
+- 免费技术支持和部署指导
+- 优先体验新功能特性
+
+## ⚙️ 高级配置
+
+### 修改关键参数
+
+#### Agent配置参数
+
+每个Agent都有专门的配置文件，可根据需求调整：
+
+```python
+# QueryEngine/utils/config.py
+class Config:
+ max_reflections = 2 # 反思轮次
+ max_search_results = 15 # 最大搜索结果数
+ max_content_length = 8000 # 最大内容长度
+
+# MediaEngine/utils/config.py
+class Config:
+ comprehensive_search_limit = 10 # 综合搜索限制
+ web_search_limit = 15 # 网页搜索限制
+
+# InsightEngine/utils/config.py
+class Config:
+ default_search_topic_globally_limit = 200 # 全局搜索限制
+ default_get_comments_limit = 500 # 评论获取限制
+ max_search_results_for_llm = 50 # 传给LLM的最大结果数
```
-#### 配置情感分析模型
+#### 情感分析模型配置
-1. **选择模型**：
- - BERT微调模型（精度高）
- - GPT-2 LoRA（速度快）
- - Qwen小模型（平衡型）
- - 机器学习基线（轻量级）
-
-2. **模型切换**：
```python
# InsightEngine/tools/sentiment_analyzer.py
-MODEL_TYPE = "bert" # 可选: "bert", "gpt2", "qwen", "ml"
+SENTIMENT_CONFIG = {
+ 'model_type': 'multilingual', # 可选: 'bert', 'multilingual', 'qwen'
+ 'confidence_threshold': 0.8, # 置信度阈值
+ 'batch_size': 32, # 批处理大小
+ 'max_sequence_length': 512, # 最大序列长度
+}
```
-#### 自定义报告模板
+### 接入不同的LLM模型
-在 `ReportEngine/report_template/` 目录下创建新模板：
+系统支持多种LLM提供商，可在各Agent的配置中切换：
+
+```python
+# 在各Engine的utils/config.py中配置
+class Config:
+ default_llm_provider = "deepseek" # 可选: "deepseek", "openai", "kimi", "gemini"
+
+ # DeepSeek配置
+ deepseek_api_key = "your_api_key"
+ deepseek_model = "deepseek-chat"
+
+ # OpenAI兼容配置
+ openai_api_key = "your_api_key"
+ openai_model = "gpt-3.5-turbo"
+ openai_base_url = "https://api.openai.com/v1"
+
+ # Kimi配置
+ kimi_api_key = "your_api_key"
+ kimi_model = "moonshot-v1-8k"
+
+ # Gemini配置
+ gemini_api_key = "your_api_key"
+ gemini_model = "gemini-pro"
+```
+
+### 更改情感分析模型
+
+系统集成了多种情感分析方法，可根据需求选择：
+
+#### 1. 基于BERT的微调模型（精度最高）
+
+```bash
+# 使用BERT中文模型
+cd SentimentAnalysisModel/WeiboSentiment_Finetuned/BertChinese-Lora
+python predict.py --text "这个产品真的很不错"
+```
+
+#### 2. GPT-2 LoRA微调模型（速度较快）
+
+```bash
+cd SentimentAnalysisModel/WeiboSentiment_Finetuned/GPT2-Lora
+python predict.py --text "今天心情不太好"
+```
+
+#### 3. 小型Qwen模型（平衡型）
+
+```bash
+cd SentimentAnalysisModel/WeiboSentiment_SmallQwen
+python predict_universal.py --text "这次活动办得很成功"
+```
+
+#### 4. 传统机器学习方法（轻量级）
+
+```bash
+cd SentimentAnalysisModel/WeiboSentiment_MachineLearning
+python predict.py --model_type "svm" --text "服务态度需要改进"
+```
+
+#### 5. 多语言情感分析（支持22种语言）
+
+```bash
+cd SentimentAnalysisModel/WeiboMultilingualSentiment
+python predict.py --text "This product is amazing!" --lang "en"
+```
+
+### 接入自定义业务数据库
+
+#### 1. 修改数据库连接配置
+
+```python
+# config.py 中添加您的业务数据库配置
+BUSINESS_DB_HOST = "your_business_db_host"
+BUSINESS_DB_PORT = 3306
+BUSINESS_DB_USER = "your_business_user"
+BUSINESS_DB_PASSWORD = "your_business_password"
+BUSINESS_DB_NAME = "your_business_database"
+```
+
+#### 2. 创建自定义数据访问工具
+
+```python
+# InsightEngine/tools/custom_db_tool.py
+class CustomBusinessDBTool:
+ """自定义业务数据库查询工具"""
+
+ def __init__(self):
+ self.connection_config = {
+ 'host': config.BUSINESS_DB_HOST,
+ 'port': config.BUSINESS_DB_PORT,
+ 'user': config.BUSINESS_DB_USER,
+ 'password': config.BUSINESS_DB_PASSWORD,
+ 'database': config.BUSINESS_DB_NAME,
+ }
+
+ def search_business_data(self, query: str, table: str):
+ """查询业务数据"""
+ # 实现您的业务逻辑
+ pass
+
+ def get_customer_feedback(self, product_id: str):
+ """获取客户反馈数据"""
+ # 实现客户反馈查询逻辑
+ pass
+```
+
+#### 3. 集成到InsightEngine
+
+```python
+# InsightEngine/agent.py 中集成自定义工具
+from .tools.custom_db_tool import CustomBusinessDBTool
+
+class DeepSearchAgent:
+ def __init__(self, config=None):
+ # ... 其他初始化代码
+ self.custom_db_tool = CustomBusinessDBTool()
+
+ def execute_custom_search(self, query: str):
+ """执行自定义业务数据搜索"""
+ return self.custom_db_tool.search_business_data(query, "your_table")
+```
+
+### 自定义报告模板
+
+#### 1. 创建模板文件
+
+在 `ReportEngine/report_template/` 目录下创建新的Markdown模板：
```markdown
-# 自定义报告模板
-## 舆情概览
-${overview}
+
+# 企业品牌舆情监测报告
-## 情感分析
-${sentiment_analysis}
+## 📊 执行摘要
+{executive_summary}
-## 关键观点
-${key_insights}
+## 🔍 品牌提及分析
+### 提及量趋势
+{mention_trend}
-## 趋势预测
-${trend_prediction}
+### 情感分布
+{sentiment_distribution}
+
+## 📈 竞品对比分析
+{competitor_analysis}
+
+## 🎯 关键观点摘要
+{key_insights}
+
+## ⚠️ 风险预警
+{risk_alerts}
+
+## 📋 改进建议
+{recommendations}
+
+---
+*报告类型：企业品牌舆情监测*
+*生成时间：{generation_time}*
+*数据来源：{data_sources}*
```
-### 监控与日志
+#### 2. 在Web界面中使用
-#### 查看系统日志
+系统支持上传自定义模板文件（.md或.txt格式），可在生成报告时选择使用。
-所有日志文件位于 `logs/` 目录：
-- `query.log`: QueryEngine运行日志
-- `media.log`: MediaEngine运行日志
-- `insight.log`: InsightEngine运行日志
-- `forum.log`: ForumEngine论坛交流日志
-- `report.log`: ReportEngine生成日志
-
-#### Agent论坛交流
-
-ForumEngine提供多Agent协作交流功能：
-1. Agent行动前读取论坛交流信息
-2. Agent思考后决定是否分享关键发现
-3. 汇总所有Agent的交流信息
-4. 为ReportEngine提供协作数据基础
-
-## 故障排除
-
-### 常见问题
-
-#### 1. 端口占用
-```bash
-# 查看端口占用（Windows）
-netstat -ano | findstr :5000
-netstat -ano | findstr :8501
-
-# 结束占用进程
-taskkill /F /PID <进程ID>
-```
-
-#### 2. 编码问题
-```python
-# 在代码开头添加
-import sys
-import os
-os.environ['PYTHONIOENCODING'] = 'utf-8'
-os.environ['PYTHONUTF8'] = '1'
-```
-
-#### 3. Playwright安装失败
-```bash
-# 手动安装
-python -m playwright install chromium --with-deps
-```
-
-#### 4. MySQL连接失败
-- 检查MySQL服务是否启动
-- 确认用户权限配置
-- 检查防火墙设置
-
-## 贡献指南
+## 🤝 贡献指南
我们欢迎所有形式的贡献！
-1. Fork项目
-2. 创建Feature分支 (`git checkout -b feature/AmazingFeature`)
-3. 提交更改 (`git commit -m 'Add some AmazingFeature'`)
-4. 推送到分支 (`git push origin feature/AmazingFeature`)
-5. 开启Pull Request
+### 如何贡献
-## 许可证
+1. **Fork项目**到您的GitHub账号
+2. **创建Feature分支**：`git checkout -b feature/AmazingFeature`
+3. **提交更改**：`git commit -m 'Add some AmazingFeature'`
+4. **推送到分支**：`git push origin feature/AmazingFeature`
+5. **开启Pull Request**
-本项目采用 MIT 许可证。详见 [LICENSE](LICENSE) 文件。
+### 贡献类型
-## 联系我们
+- 🐛 Bug修复
+- ✨ 新功能开发
+- 📚 文档完善
+- 🎨 UI/UX改进
+- ⚡ 性能优化
+- 🧪 测试用例添加
-- 项目地址：[https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem)
-- 邮箱：670939375@qq.com
-- Issues：[项目Issues](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/issues)
+### 开发规范
+
+- 代码遵循PEP8规范
+- 提交信息使用清晰的中英文描述
+- 新功能需要包含相应的测试用例
+- 更新相关文档
+
+## 📄 许可证
+
+本项目采用 [MIT许可证](LICENSE)。详细信息请参阅LICENSE文件。
+
+## 🎉 支持与联系
+
+### 获取帮助
+
+- **项目主页**：[GitHub仓库](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem)
+- **问题反馈**：[Issues页面](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/issues)
+- **功能建议**：[Discussions页面](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/discussions)
+
+### 联系方式
+
+- 📧 **邮箱**：670939375@qq.com
+- 💬 **QQ群**：[加入技术交流群]
+- 🐦 **微信**：[扫码添加技术支持]
+
+### 商务合作
+
+- 🏢 **企业定制开发**
+- 📊 **大数据服务**
+- 🎓 **学术合作**
+- 💼 **技术培训**
+
+### 云服务申请
+
+**免费云数据库服务申请**：
+📧 发送邮件至：670939375@qq.com
+📝 标题：微博舆情云数据库申请
+📝 说明：您的使用场景和需求
+
+## 👥 贡献者
+
+感谢以下优秀的贡献者们：
+
+[](https://github.com/666ghj/Weibo_PublicOpinion_AnalysisSystem/graphs/contributors)
+
+---
+
+
+
+**⭐ 如果这个项目对您有帮助，请给我们一个星标！**
+
+Made with ❤️ by [微博舆情分析团队](https://github.com/666ghj)
+
+
diff --git a/static/image/banner_compressed.png b/static/image/banner_compressed.png
index 2037fd5..c4d2e58 100644
Binary files a/static/image/banner_compressed.png and b/static/image/banner_compressed.png differ
diff --git a/static/image/logo_mindspider.png b/static/image/logo_mindspider.png
new file mode 100644
index 0000000..e50ffa3
Binary files /dev/null and b/static/image/logo_mindspider.png differ
diff --git a/static/image/system_schematic.png b/static/image/system_schematic.png
new file mode 100644
index 0000000..9db2905
Binary files /dev/null and b/static/image/system_schematic.png differ