Charity Owusua Jnr Danso
Research Subjects
Charity Owusua Jnr Danso has a total of 2 co-authors. Here are the top co-authors by number of shared publications:
- Jeffrey Nii Armah Aryee
Kwame Nkrumah University of Science and Technology
1 shared publication - Linus Kweku Labik
Department of Physics, Kwame Nkrumah University of Science and Technology, University Post Office, Private Mail Bag, Kumasi, Ghana
1 shared publication
Latest Publications
- Open Access
- 1 January 2025
- Published by Elsevier in SSRN Electronic Journal
Utilizing Machine Learning Techniques to Enhance Climate-Crop Yield Nexus in the Akyemansa District of Ghana
Agriculture is central to Africa's economy, contributing significantly to poverty reduction and aligning with the UN's SDGs. In Ghana, it contributes 40% of GDP, 75% of export earnings, and employs 55% of the workforce. However, climate change threatens agriculture, particularly rain-dependent smallholder farms. This study assesses the impact of climate variability on crop yield in the Akyemansa District of Ghana and employs machine learning (ML) techniques to enhance food security. Stakeholders expressed concerns over unpredictability of rainfall patterns, rising temperatures, pests and diseases outbreaks, all affecting cocoa yield. Statistical analyses revealed strong correlations between dry days, rainy days, precipitation, and wind speed and crop productivity. ML models such as Multi-linear Regression (LR), Random Forest (RF), Decision Tree (DT), and Support Vector Regression (SVR) techniques were evaluated using mean absolute error (MAE) and R-squared (R²) score. RF and SVR consistently outperformed LR and DT. RF achieved the highest R² scores (68–74%) and lowest MAE (43–53%), followed by SVR. LR and DT had lower R² scores and higher MAE, with DT performing worse Notably, RF and SVR accuracy improved with more training years. Despite small datasets, RF and SVR demonstrated strong predictive capabilities, showing their potential to improve agricultural yield predictions under changing climatic conditions.
- 1 January 2025
- Published by Elsevier in SSRN Electronic Journal
Open AccessUtilizing Machine Learning Techniques to Enhance Climate-Crop Yield Nexus in the Akyemansa District of Ghana
Agriculture is central to Africa's economy, contributing significantly to poverty reduction and aligning with the UN's SDGs. In Ghana, it contributes 40% of GDP, 75% of export earnings, and employs 55% of the workforce. However, climate change threatens agriculture, particularly rain-dependent smallholder farms. This study assesses the impact of climate variability on crop yield in the Akyemansa District of Ghana and employs machine learning (ML) techniques to enhance food security. Stakeholders expressed concerns over unpredictability of rainfall patterns, rising temperatures, pests and diseases outbreaks, all affecting cocoa yield. Statistical analyses revealed strong correlations between dry days, rainy days, precipitation, and wind speed and crop productivity. ML models such as Multi-linear Regression (LR), Random Forest (RF), Decision Tree (DT), and Support Vector Regression (SVR) techniques were evaluated using mean absolute error (MAE) and R-squared (R²) score. RF and SVR consistently outperformed LR and DT. RF achieved the highest R² scores (68–74%) and lowest MAE (43–53%), followed by SVR. LR and DT had lower R² scores and higher MAE, with DT performing worse Notably, RF and SVR accuracy improved with more training years. Despite small datasets, RF and SVR demonstrated strong predictive capabilities, showing their potential to improve agricultural yield predictions under changing climatic conditions.