|
|
--- |
|
|
title: RICA - AI Revenue Intelligence Agent |
|
|
emoji: 🤖 |
|
|
colorFrom: blue |
|
|
colorTo: purple |
|
|
sdk: streamlit |
|
|
sdk_version: 1.28.0 |
|
|
app_file: app.py |
|
|
pinned: false |
|
|
license: mit |
|
|
python_version: "3.10" |
|
|
--- |
|
|
|
|
|
# 🤖 RICA - Revenue Intelligence & Customer Analytics Agent |
|
|
|
|
|
Advanced AI agent for autonomous business intelligence using SAP data and machine learning. |
|
|
|
|
|
## Features |
|
|
|
|
|
- 🧠 **AI-Powered Churn Prediction**: Machine learning models predict customer churn risk |
|
|
- 📊 **Real-time Data Analysis**: Direct analysis of SAP data structures |
|
|
- 🌐 **Market Intelligence**: External data integration and competitive insights |
|
|
- 🤖 **Autonomous Agent**: LLM-powered decision making and recommendations |
|
|
- 📈 **Business Impact**: Actionable insights for revenue optimization |
|
|
|
|
|
## Technology Stack |
|
|
|
|
|
- **Agent Framework**: smolagents with OpenAI GPT |
|
|
- **ML Models**: Scikit-learn with automated training |
|
|
- **Data Processing**: DuckDB for high-performance analytics |
|
|
- **Real SAP Data**: SAP/SALT dataset from Hugging Face Hub |
|
|
- **UI**: Streamlit for interactive experience |
|
|
|
|
|
## Usage |
|
|
|
|
|
1. Enter your OpenAI API key in the sidebar |
|
|
2. Select analysis type (Comprehensive Review, Churn Analysis, etc.) |
|
|
3. Click "Run Analysis" to execute AI-powered insights |
|
|
4. Review recommendations and take action |
|
|
|
|
|
## Model Training |
|
|
|
|
|
The churn prediction model is automatically trained on first use using real SAP customer and sales data. Training typically takes 1-2 minutes and creates a persistent model for future predictions. |
|
|
|
|
|
## Data Sources |
|
|
|
|
|
- **Customer Data**: I_Customer from SAP/SALT dataset |
|
|
- **Sales Data**: I_SalesDocument and I_SalesDocumentItem |
|
|
- **External Data**: Industry news and competitive intelligence |
|
|
|
|
|
Built with ❤️ for enterprise AI and business intelligence. |
|
|
|
