SheilaMumbi/Regional_Crop_Yield_Prediction_Kenya

Developing a supervised machine learning model capable of predicting regional crop yield in Kenya using historical production data, climate variables, and agricultural input usage. # Kenya Regional Crop Yield Prediction Using Machine Learning ## 📌 Project Overview Agriculture is a critical sector in Kenya, supporting food security, employment, and economic stability. However, crop yields are highly sensitive to climate variability, regional differences, and agricultural input usage. This project develops a machine learning pipeline to predict regional crop yields using historical production data, climate indicators, and pesticide usage. The goal is to provide data-driven insights that can support early warning systems, resource allocation, and agricultural planning. --- ## 🎯 Problem Statement Traditional yield estimation relies on historical averages, limiting the ability to anticipate food shortages and climate-related production risks. This project addresses the question: > **How accurately can regional crop yields in Kenya be predicted using climate variables and agricultural input data?** --- ## 🧭 Objectives - Predict crop yield at the regional level using supervised machine learning. - Identify regions most sensitive to rainfall and temperature variability. - Evaluate the influence of pesticide usage on yield outcomes. - Explore whether predictive models can provide early warning signals for potential food shortages. --- ## 🗂️ Data Sources ### 1. Kenya Agricultural Production Dataset - **Source:** Kaggle - **File:** `Kenya Agricultural production.xlsx` - **Variables:** Year, Crop, Area Harvested, Production, Yield - **Purpose:** Historical crop production metrics. ### 2. HarvestStat Africa – Regional Crop Data - **Source:** HarvestStat GitHub - **File:** `adm_crop_production_KE.csv` - **Variables:** Region, Crop, Area, Production - **Purpose:** Regional yield modeling. ### 3. Climate Data (Rainfall & Temperature) - **Source:** OpenAfrica - **Variables:** Year, Month, Rainfall (mm), Temperature (°C) - **Purpose:** Environmental drivers of yield variability. ### 4. Pesticide Usage Data - **Source:** KAPSARC Data Portal - **Va …