LeiteIntel is a data analysis system focused on the milk production chain in Brazil, designed to support rural producers, technicians, and researchers in making strategic decisions. Using advanced data analysis techniques, the project allows exploration of trends, forecasting, and generation of customized reports on milk production across different states and periods.

The system is built following the CRISP-DM (Cross-Industry Standard Process for Data Mining) model, a well-established framework for conducting data mining and analysis projects in an organized and efficient manner. This process includes the stages of business understanding, data understanding, data preparation, modeling, evaluation, and deployment.

Additionally, the project works with Big Data concepts, handling expanded and complex datasets to extract valuable insights, focusing on production, price, climate, and technology applied to dairy farming.

• Interactive data visualization by state and year, with detailed milk production charts.
• Statistical analyses and summarized tables showing averages by production type and technology use.
• Prediction model to estimate future production based on climatic and economic variables.
• Personalized PDF report generation, including interpretative texts and graphs/tables, with user customization options.
• Modular structure facilitating maintenance and system expansion.

• Python with Streamlit for building the interactive web interface.
• Pandas, Seaborn, and Matplotlib for data manipulation and visualization.
• Scikit-learn for predictive modeling using linear regression and preprocessing.
• FPDF for generating PDF reports.
• Clear modular organization to ease development and reuse.

1. Make sure the data file leiteintel_base_ampliada.csv is in the data/ folder.
2. Run the app with: streamlit run app.py
3. At startup, the system checks if the cleaned dataset exists; if not, it automatically generates the processed version.
4. Use the sidebar to select state, year, and type of analysis.
5. Explore the charts, tables, and make predictions.
6. Customize and generate complete PDF reports directly from the app.

For questions or contributions, please contact via GitHub or email: karlinharural@gmail.com.

Thank you for using LeiteIntel!
Innovation in milk production starts with well-analyzed data.