Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

EdnaAnnQuarshie/Predicting-Cocoa-Seedling-Survival-in-Ghana-using-Machine-Learning-and-Early-Growth-Indicators

Domain:

agriculture

Record type:

project
Creator:
Edn
Host:
Predicting Cocoa Seedling Survival in Ghana using Machine Learning and Early Growth Indicators ## Why this project Ghana is the world's second largest cocoa producer, yet the sector is held back by persistently high seedling mortality after transplanting. As an extension officer at Ghana Cocoa Board (COCOBOD), I worked directly with farmers who raised the same concern again and again: the seedlings they collected from cocoa stations often did not survive in the field. This project set out to investigate the problem using data, and to test whether seedling survival could be predicted early enough for nurseries and extension officers to act. ## The question Can early growth indicators, measured at the nursery stage, predict which cocoa seedlings will survive after transplanting to the field? ## Data The study used records for 1,000 seedlings from COCOBOD's Akwadum nursery, tracked across four early growth indicators during the nursery phase. The dataset was collected and prepared for analysis as part of my MSc Business Analytics dissertation at the University of Northampton. ## Methods - Exploratory data analysis and data cleaning in Python (Pandas, NumPy) - Class imbalance handled using SMOTE - Classification models: Random Forest and Logistic Regression - Cross-validation and performance evaluation using accuracy, precision, recall and F1-score - Feature importance analysis to identify which early growth indicators carried the most predictive weight ## Key findings - The nursery itself achieved a **98.6% seedling survival rate**, which reframes Ghana's cocoa mortality issue. The problem is not nursery quality, it is field establishment after transplanting. - Early growth indicators showed meaningful predictive signal, suggesting that nursery-stage monitoring can flag at-risk seedlings before they reach farmers. - Classification accuracy of over 92% was achieved using Random Forest with SMOTE. ## Why it matters If nurseries and extension officers can ide …

Visit

github.com

Similar

Predicting South African GDP growth using machine learning approachesPredicting House Rental Prices in Ghana Using Machine LearningEarly Childhood Anemia in Ghana: Prevalence and Predictors Using Machine Learning TechniquesEarly Detection of Family Violence Risk Using Ensemble Machine Learning on Psychosocial and Demographic IndicatorsComparative performance of machine learning models in forecasting Somalia’s GDP growth using macroeconomic and environmental indicatorsPredicting the Performance of Rural Banks in Ghana Using Machine Learning Approach

Predicting South African GDP growth using machine learning approaches

This study uses secondary data from Stats SA (

Predicting House Rental Prices in Ghana Using Machine Learning

This study investigates the efficacy of machine learning models for predicting house rental prices i

Early Childhood Anemia in Ghana: Prevalence and Predictors Using Machine Learning Techniques

Background/Objectives: Early childhood anemia is a severe public health concern and the most common

Early Detection of Family Violence Risk Using Ensemble Machine Learning on Psychosocial and Demographic Indicators

Objective: The objective of this study was to develop and validate an ensemble machine learning fram

Comparative performance of machine learning models in forecasting Somalia’s GDP growth using macroeconomic and environmental indicators

Abstract This study evaluates the comparative performance o

Predicting the Performance of Rural Banks in Ghana Using Machine Learning Approach

The idea of rural banks was introduced as a result of limited commercial bank branches in rural area