Meeting Analysis Dashboard

This Streamlit application analyzes meeting transcripts to provide insights into energy levels, social dynamics, mood patterns, and topic distribution.

Features

• Upload and analyze meeting transcripts
• Advanced text analysis using Anthropic's Claude API
• Visualize energy level distribution
• Track social interaction patterns
• Monitor mood trends
• Analyze topic distribution
• Download analysis results

Setup

1. Install the required dependencies:

pip install -r requirements.txt

2. Set up your Anthropic API key:

   • Copy the .env.example file to .env
   • Replace your_api_key_here with your actual Anthropic API key
   • If no API key is provided, the system will fall back to basic pattern matching

3. Run the Streamlit app:

streamlit run app.py

Usage

1. The app will automatically load any .txt files from the raw_data directory
2. The system will use Anthropic's Claude API for advanced text analysis (if API key is provided)
3. Navigate through the different tabs to view various analyses:
   • Energy Levels: Distribution of energy levels throughout the meeting
   • Social Dynamics: Analysis of positive and negative interactions
   • Mood Analysis: Distribution of different moods
   • Topic Analysis: Distribution of discussed topics
4. Use the "View Raw Analysis Data" expander to see the detailed analysis results
5. Download the analysis results using the download buttons

Analysis Methods

The system uses two methods of analysis:

1. Anthropic Claude API (Primary)

   • Uses advanced language understanding for better context awareness
   • More accurate identification of subtle emotional indicators
   • Better handling of complex social interactions
   • Requires API key

2. Pattern Matching (Fallback)

   • Uses regular expressions to identify key patterns
   • Works without external API dependencies
   • More limited in understanding context
   • Serves as a backup when API is unavailable

Data Structure

The application analyzes text files for:

• Energy levels (High, Medium, Low)
• Social interactions (Positive, Negative)
• Moods (Happy, Sad, Anxious, etc.)
• Topics (Work, Social, Personal, etc.)

Contributing

Feel free to submit issues and enhancement requests!