Update README.md

README.md

@@ -1,43 +1,73 @@
 ---
-title: RAG Document Processing
-emoji: 📄
+title: Email Thread Analyzer with AI Q&A
+emoji: 📧
 colorFrom: blue
-colorTo: green
-sdk: docker
-app_port: 7860
-pinned: false
-models:
-  - sentence-transformers/all-MiniLM-L6-v2
-  - facebook/bart-large-cnn
-license: apache-2.0
+colorTo: indigo
+sdk: gradio
+sdk_version: 4.44.0
+app_file: app.py
+pinned: true
+license: mit
 ---
 
-# RAG Document Processing
+# Email Thread Analyzer with AI Q&A
 
-This Hugging Face Space contains a RAG (Retrieval Augmented Generation) system for processing documents, built entirely with Hugging Face tools and models.
+An AI-powered tool for analyzing email threads, searching for relevant conversations, and generating Q&A pairs based on email content.
 
 ## Features
 
-- Automated processing of text files
-- OCR and document processing with HF models
-- Vector storage with FAISS
-- Embeddings with sentence-transformers
-- Simple query interface with Gradio
+- Microsoft Graph API integration for email access
+- Email thread organization and visualization
+- Semantic search for finding relevant threads
+- AI-powered Q&A generation based on thread content
+- Data visualization for email analytics
+- Export functionality for thread data and Q&A
 
-## Implementation
+## Authentication
 
-This is a Hugging Face-native implementation of the original cloud-based RAG system. It uses:
-
-- Sentence Transformers for embeddings
-- FAISS for vector search
-- Transformers models for document processing
-- Persistent storage for data
-- A100 GPU acceleration (for pro users)
+1. Click on the "Authentication" tab
+2. Click "Get Authentication URL" button
+3. Open the URL in a new browser tab and authenticate with Microsoft
+4. After successful authentication, you'll be redirected. Copy the authorization code from the URL
+5. Paste the code in the "Authorization Code" field and click "Submit"
 
 ## Usage
 
-1. Upload documents to be processed
-2. The system automatically processes and indexes them
-3. Query the processed documents through the interface
+### Email Extraction
+
+1. Select a mail folder from the dropdown
+2. Configure extraction parameters (max emails, batch size, date range)
+3. Click "Extract Emails" to fetch and organize email threads
+
+### Thread Analysis
+
+1. Click on the "Thread Analysis" tab
+2. Click "Generate Analytics" to visualize email metrics
+3. Explore the thread size distribution, activity timeline, and participant charts
+
+### Search
+
+1. Click on the "Search" tab
+2. Enter a search query and click "Search"
+3. Select a thread from the results and click "View Thread" to see details
+
+### Q&A
+
+1. Select a thread using the Search tab
+2. Click "Generate Q&A" to create AI-powered questions and answers
+3. Explore the generated Q&A pairs and summary
+4. Export the thread data with Q&A for further analysis
+
+## Privacy & Security
+
+- All processing happens within the application
+- Authentication is handled securely through Microsoft's OAuth2 flow
+- No email data is stored persistently outside of your session
+
+## Technical Details
 
-See the CLAUDE_HF.md file for implementation details.
+This application uses:
+- MSAL for Microsoft authentication
+- Sentence-Transformers for semantic search
+- Transformers for question answering
+- Plotly for data visualization