DATS-group/Coffee_Prediction_Analysis

This is a python group project whereby we shall be taking a Kenya coffee produce expectancy dataset, cleaning the dataset, working on it then creating a model using various libraries AI Analysis and Prediction of Kenyan Coffee Exports: Formal Project Report 1. Problem Definition and Objective Kenya produces some of the most premium Arabica coffee globally, yet the structural and regulatory hurdles within its export pipeline remain formidable. The supply chain is characterized by extreme fragmentation, with approximately 70% of production originating from over 700,000 smallholder farmers. These producers must navigate complex cooperative societies and multiple intermediaries, often resulting in delayed payments and diluted earnings. Furthermore, the industry faces significant climate volatility and the looming pressure of the European Union Deforestation Regulation (EUDR). Compliance with the EUDR requires expensive digital traceability and GPS geo-mapping—a substantial investment for a market that consumes nearly 60% of Kenya's coffee.The primary objective of this project is to develop a robust predictive model to assist smallholder farmers and cooperatives in monitoring price trends and planning financial payouts. By leveraging historical trade and macroeconomic data, we aim to bridge the gap between global market volatility and rural economic stability. This project is formally classified as a Regression task, focused on predicting export values and price fluctuations to support data-driven decision-making. 2. Data Acquisition The study utilized a longitudinal dataset curated from Kenyan government portals, ensuring the use of authoritative economic and agricultural records.Technical Specifications| Detail | Specification || ------ | ------ || Time Span | 2001–2020 (Approximately 19 years) || Size | 21 rows and 29 initial columns || Data Format | CSV || Key Features | Import prices (South Korea, Germany, USA, Belgium), Average prices (Kenya/Foreign), Exchange rates (Real/Nominal), Macroeconomic indicators (GDP growth, Population growth, Interest rates), and Exporter-specific volumes. | 3. Data Cleaning and Preprocessing The raw dataset require …