Crop Yield Forecasting
As part of my Master's thesis, I developed a machine learning model to forecast crop yields in Mexico using weather and historical data. The model aimed to enhance yield prediction accuracy, providing valuable insights for farmers and businesses to make data-driven, proactive decisions.
Features
- Extracted and processed data using Python from multiple sources, including Excel files from government platforms, a web crawler for historical hurricane data from NOAA, and meteorological data via the OpenMeteo API.
- Developed an interactive Power BI dashboard to analyze and visualize vegetable production trends across Mexico.
- Prepared training and test datasets for predictive modeling, ensuring quality and consistency for analysis.
- Built intermittent models in Python and utilized IBM Watson Machine Learning to predict hurricane and drought frequencies.
- Visualized Root Mean Square Error (RMSE) of predictive models to determine the most accurate model.
- Used Matplotlib to visualize feature importance in predictive models for better insights.
- Developed machine learning models with Watson IBM to forecast vegetable production based on meteorological phenomenon predictions, integrating climate event forecasts with agricultural outputs.
Web Crawler for Data Extraction Automation
Developed a Python-based web crawler utilizing Selenium to extract historical hurricane data, automating the data extraction process to retrieve information for predictive modeling.
Web Crawler for Data Extraction Automation
Developed a Python-based web crawler utilizing Selenium to extract historical hurricane data, automating the data extraction process to retrieve information for predictive modeling.
