Analysis of WeFarm farmer questions and World Bank climate data in Kenya. Uses Python, NLP, and LDA to explore how rainfall and temperature influence farmer behavior, revealing seasonal trends in topics like pests, irrigation, and soil management for data-driven agricultural support.
# 🌾 Kenya Climate–WeFarm Analysis (DataKind Volunteer Project)
This repository explores the relationship between **climate conditions** and **farmer communication behavior** using data from **WeFarm** and the **World Bank Climate Knowledge Portal**.
The goal is to understand how rainfall and temperature patterns influence the types and frequency of farmer questions.
---
## 📊 Project Overview
This analysis focuses on smallholder farmers in Kenya and examines:
- How monthly **rainfall** and **temperature** affect farmer engagement.
- What **topics** farmers discuss most during specific seasons (pests, irrigation, soil, etc.).
- How **data science** and **NLP** can guide better agricultural decision-making.
---
## 🔍 Methodology
- **Data Cleaning & Integration:** Combined World Bank climate data with WeFarm questions.
- **Exploratory Data Analysis:** Monthly trends, rainfall correlations.
- **Topic Modeling (LDA):** Identified discussion themes automatically from question text.
- **Visualization:** Showed seasonal topic variation and weather–behavior correlations.
---
## đź§ Key Insights
- Rainy seasons bring spikes in pest and planting-related questions.
- Dry periods shift interest toward irrigation and soil management.
- Farmers’ digital behavior can predict upcoming agricultural challenges.
---
## ⚙️ Tools & Libraries
Python · Pandas · NumPy · Matplotlib · Scikit-learn · LDA (NLP)
---
## đź§© Data Sources
- World Bank Climate Knowledge Portal
- WeFarm Farmer Dataset (via DataKind / Producers Direct)
---
## 🧑‍💻 Author
**Karam Alojaili**
DataKind Volunteer | Data for Social Impact | Bioinformatics & Climate Analytics
---
## đź“‚ How to Run
Open the notebook on Google Colab or Jupyter:
1. Upload `Kenya_Climate_WeFarm_KaramAlojaili.ipynb`
2. Run cells sequentially.
3. Outputs include CSVs and PNG visualizations.