bizzaccelerator/corn-yield-prediction

Machine learning system to predict household-level corn production in Kenya, supporting food security decisions and resource allocation. # Corn yield prediction service in Kenia *A Python-based predictive application for estimating corn yields using survey data, containerized with Docker.* This fictional project was originally developed as a midterm evaluation for the Machine Learning Zoomcamp and improved as a final projecto for MLops Zoomcamp, both offered by Data Talks Club. Method and objectives were defined for educational purposes only, so I can show the knowledge appropiated during the mentioned training. The current project simulates a real scenario of information gathering to support effective political decision-making in a mayor's office in Kenya, aiming to ensure food security in the region. Photo: ©somkak – stock.adobe.com ## Problem statement This could be understood in two leves: a business problem and a technical problem. ### _Business problem:_ Certain region in Kenya has experienced rapid population growth over the past decade in an underdeveloped economic environment. The social group living in this region considers _corn_ as the preferred base for most typical dishes; however, the low level of precipitation threatens sufficient production in the coming years. The Mayor's Office seeks to make the best decisions to ensure food security in the county. To acheive that goal, the prediction of corn production at a household level is a must. That’s why the managing team at the political office needs to know the expected levels of corn production at a household level, the key variables affecting it, so they can further improve the resources allocation process. ### _Technical problem:_ The county in Kenya is facing a potential food security risk due to rapid population growth, underdeveloped economic conditions, and declining precipitation levels. Since corn is the staple food in this region, the Mayor’s Office requires a reliable and scalable system to predict household-level corn production and identify the key variables influencing yield. These insights are essential for making d …