All theses

Permanent URI for this collection

http://62.72.18.19:4000/handle/123456789/6

Browse

Now showing 1 - 18 of 18

Evaluation of Potential Impacts of the Introduction of Electric Vehicles in the Transportation Sector in Ghana
(UENR, 2025-09) Arthur, O.J.
This study evaluated the potential environmental, economic, and socio-technical impacts of adopting electric vehicles (EVs) in Ghana using well-to-wheels analysis, cost comparison, and policy review. Findings reveal that substituting internal combustion engine vehicles (ICEVs) with EVs could reduce lifecycle CO₂ emissions by 30–45%, equivalent to 2.5–3.2 tonnes of CO₂ avoided per vehicle annually. Local air pollutants such as NOₓ and PM₂.₅ could decline by 25– 40% and up to 30%, respectively, improving air quality and public health. Although EVs currently cost 30–50% more than ICEVs, their lower operating costs yield annual savings of US$800–1,200, translating into lifetime savings of US$8,000–12,000 per vehicle. These benefits make EVs economically viable in the long term, especially with supportive government incentives. Policy implications emphasize the need for a coherent national EV framework integrating fiscal incentives, renewable-powered charging infrastructure, and public–private partnerships. Policymakers should prioritize EV integration into Ghana’s transport and energy systems through tax reforms, import duty waivers, and local manufacturing support. Electrifying commercial transport fleets could further enhance emission reductions and energy security. The study contributes to existing literature by providing context-specific empirical evidence on EV adoption in Ghana, addressing gaps in environmental, economic, and policy analysis within developing economies. It extends the Diffusion of Innovations and Sustainable Development theories to assess socio-technical readiness and institutional feasibility in low- and middle-income contexts. Overall, EV adoption represents a viable pathway toward Ghana’s sustainable transport transition and national emission reduction goals.
Assessing the Socioeconomic Environmental and Water Quality Impacts of Illegal Gold Mining in Mankranso Ashanti Region – Ghana - Implications for Sustainable Environmental Management
(UENR, 2025-11) Afi, K.E.
Illegal gold mining (galamsey) has become one of the most pressing environmental and social challenges in Ghana, contributing to water pollution, land degradation, and community vulnerability. Despite its role in rural livelihoods, its unregulated nature raises major concerns for sustainability. This study examined the socioeconomic and environmental impacts of illegal smallscale gold mining (galamsey) in Mankranso, Ghana. Data were collected through household surveys (n = 372), geospatial analysis of land use and land cover (LULC) change, and river water quality monitoring. Water samples were collected from upstream, midstream, and downstream sections of mining sites and analysed for heavy metals and physicochemical parameters using standard procedures. Results showed that unemployment (44.1%) and high profitability (30.9%) were the main drivers of participation in galamsey. Communities reported social challenges such as teenage pregnancy, child labour, drug abuse, and prostitution linked to mining. Water quality deteriorated significantly downstream, with turbidity, electrical conductivity, and total dissolved solids exceeding WHO standards, while heavy metals (Pb, Hg, Cd, As, Fe) surpassed permissible limits, particularly in the wet season. Cyanide was detectable but within WHO standards. LULC analysis revealed major declines in forest and agricultural land, with increases in bare land and mining areas, confirming widespread land degradation and riverbank erosion. Classification accuracies exceeded 85% (Kappa > 0.80). Overall, the findings highlight the dual role of galamsey as a livelihood source and a driver of severe environmental degradation and social disruption. The study recommends integrated interventions combining alternative livelihood support, stricter regulation, improved water supply, community education, and sustained geospatial monitoring to safeguard public health and natural resources.
Phytodisinfection of Wastewater - Efficacy of Ocimum Gratissimum Leaf Extract
(UENR, 2025-10) Agyei, A.M.
Waterborne microbial pathogens continue to pose a significant public health challenge, especially in resource-limited settings. While plant-based disinfectants offer a promising, sustainable alternative, their direct application in complex water treatment remains underresearched. This study evaluated the antimicrobial efficacy of ethanolic Ocimum gratissimum leaf extract against Escherichia coli, a primary indicator of microbial contamination. The extract’s phytochemical profile was characterized using Fourier Transform Infrared (FTIR) spectroscopy to identify key functional groups. Antimicrobial activity was initially confirmed via a disc diffusion assay, followed by quantitative determination of the Minimum Inhibitory Concentration (MIC) and Minimum Bactericidal Concentration (MBC) using broth microdilution. The extract’s dose-dependent efficacy was then assessed in both simulated contaminated water and real waste water samples. The results showed that O. gratissimum extract exhibited significant inhibitory and bactericidal activity against E. coli. In sterile contaminated water, a concentration of 50% (v/v) achieved a ≥99.9% microbial reduction, confirming its potential as a potent disinfectant. However, its effectiveness was reduced in real wastewater, where a 50% concentration achieved approximately 71% microbial reduction. This reduced efficacy is attributed to the high-strength wastewater matrix, characterized by high levels of turbidity (800 NTU) and total suspended solids (TSS) (1,400 mg/L). The study's findings confirm the extract's strong dose-dependent antimicrobial potential and its value as a complementary, plant-based disinfectant for decentralized water treatment. This research provides novel insights into the practical application of O. gratissimum, suggesting its potential as a sustainable solution in resource constrained areas.
Spatial Modeling of Microplastic Concentration in Lake Volta
(UENR, 2025-09) Andoh, N.C.
Microplastics (MPs), defined as plastic particles smaller than 5 mm, have emerged over the past decade as one of the most pressing global environmental challenges, particularly in aquatic systems. While initial research has predominantly focused on marine environments, examining MPs in sediments, surface waters, and biota, it is now increasingly evident that freshwater ecosystems are equally vulnerable. MPs enter aquatic systems through land-based sources transported via rivers and streams, as well as contributions from fisheries and shipping activities. Although global attention to MPs in freshwater is expanding, significant research gaps persist in Africa, with Ghana's inland waters—especially Lake Volta—receiving minimal investigation despite their ecological and socioeconomic importance. Specifically, there is a lack of spatially explicit data on MP distribution, limited understanding of source-to-sink pathways in large tropical reservoirs, and inadequate integration of social practices with environmental contamination patterns in the region. This study addresses these gaps through an innovative, multi-matrix spatial assessment that integrates household surveys, environmental sampling, and biological analysis within a unified analytical framework. Guided by the Pressure-State-Response (PSR) theoretical model, the research systematically examines human-induced pressures (solid waste management practices), environmental state (MP distribution in water and sediment), and biological impacts (bioaccumulation in fish), culminating in evidence-based response strategies. The study is further structured by a novel conceptual framework that maps the complete pollution pathway from land-based activities through aquatic transport to biological uptake, enabling holistic source attribution and risk assessment. Key innovations include: (1) the first comprehensive spatial mapping of MPs across Lake Volta using GISbased interpolation to identify pollution hotspots; (2) integrated sourceto-impact analysis linking household waste practices directly to environmental contamination and biological exposure; and (3) application of advanced spectroscopic techniques (ATR-FTIR) combined with ecological risk indices to quantify both immediate and long-term contamination threats. Empirical findings revealed that solid waste mismanagement—a critical pressure—was widespread, with 40.0% of lakeshore households generating predominantly plastic waste, yet only 40.0% aware of waste segregation and 51.9% disposing of waste near their homes. Environmental analyses confirmed significant MP contamination, with average concentrations of 15.88 ± 10.69 MPs/L in surface water and 148.33 ± 119.35 MPs/kg in sediment, significantly higher in sediments (p < 0.001). Fibers and polyethylene dominated, with spatial hotspots identified in fishing-intensive zones (SII, SVI, SVII). Biological uptake was confirmed with 229 MPs detected in 96 fish specimens, averaging 2.47 ± 1.30 MPs per tilapia and 2.29 ± 1.73 MPs per catfish. MPs were negatively correlated with pH and dissolved oxygen, revealing physicochemical controls on distribution. While Ecological Risk Index (ERI) values (30.52 water, 27.44 sediment) indicated low immediate risk, Polymer Hazard Index (PHI) values (14.76 water, 13.02 sediment) signaled significant long-term contamination potential. Collectively, these findings demonstrate that Lake Volta faces escalating MP pollution with clear ecological and socioeconomic implications. The study provides the first spatially explicit evidence base for MP contamination in Ghana's largest freshwater reservoir, offering novel insights into pollution drivers, distribution patterns, and biological transfer. Informed by the PSR framework, urgent interventions—including strengthened waste management, targeted plastic recovery, community education, and spatially explicit monitoring—are required to safeguard fisheries, livelihoods, and water security in the Lake Volta basin. This research establishes a transferable methodological framework for assessing MP pollution in similar freshwater systems across West Africa.
Effects of Biochar Application Rates on Growth of Terminalia Superba Seedlings Soil Chemical Properties and Soil Organic Carbon under Nursery Conditions
(UENR, 2025-09) Abonkra, Y.B.
Producing high quality seedlings is essential for successful plantation establishment and the choice of growing medium plays a critical role. This study determined the effects of biochar application rates by volume (v) on the growth of Terminalia superba seedlings and on key soil chemical properties under nursery conditions. Biochar was amended to sandy loam soil at different rates in two nursery experiments conducted. Experiment one constituted 10% (v), 25% (v) and 50% (v) biochar amendment rate and control (unamended topsoil), each replicated six times. Experiment two constituted 2% (v), 5% (v) and 10% (v) and each replicated three times. Biochar treatments in both experiments were arranged in a Randomized Complete Block Design (RCBD). In both experiments, seedling growth parameters (seedling height, root collar diameter and number of leaves) and soil chemical properties (pH, nitrogen, phosphorus, potassium and soil organic carbon) were observed. The results showed that biochar amendment significantly influenced seedling growth parameters and soil chemical properties. In experiment one, the 10% (v) biochar treatment significantly increased seedling growth parameters compared to the control, while the highest rate biochar amendment 50% (v) suppressed seedling growth parameters. Similarly, in experiment two, low biochar amendment rates 2% (v) and 5% (v) enhanced seedling growth parameters. In both experiments, biochar consistently improved soil chemical properties. In experiment one, pH increased significantly by 0.92 units, N by 164%, P by 51%, K by 68% and SOC by 114% relative to the control. In experiment two, N increased significantly by 189%, P by 61%, K by 48% and SOC by 146% compared to the control. Higher biochar amendment rates (v) significantly increased pH, N, K and SOC, while P significantly increased mainly at low to moderate rates (v). The study concludes that low biochar amendment rates (v) are optimal for raising Terminalia superba seedlings, whereas higher rates (v) are more suitable for enhancing soil carbon sequestration and long-term soil fertility.
Influence of Seasonal Climate Variabilities and Phenology of Mangifera Indica on the Population Dynamics of Diptera Tephritidae in the Nadowli-Kaleo District
(UENR, 2025-12) Ruhullah, A.B.
Destruction of Diptera tephritidae has imposed quarantine restrictions and rapid economic losses in the horticulture industry. The nexus between agricultural productivity and climate change has garnered growing attention that requires scientific investigation. The study aims to examine the interactions between seasonal climate variability and the phenology of Mangifera indica on the population dynamics of Diptera tephritidae, anchored in climate ecology theory, phenological theory, population dynamics theory, and a systems ecological framework. Randomized Complete Block Design was adopted for the field layout, and Longitudinal studies were employed in gathering past and present data of fruit fly visitation and phenology of Mangifera indica, over a period of four production seasons. Climatic data were sourced from Ghana Meteorological Agency data base. Data analysis involved a generalized linear model and a Seasonal Harmonic Analysis. ANOVA test and multiple regression were performed, and a Random Forest regressor model was used for prediction. The model was trained on data from 2022 to 2024 and then tested to predict the fly count for the entire 2025. The study found a strong seasonal pattern in the fruit fly population. Total monthly rainfall and minimum temperature proved to be the most influential factors. In particular, rainfall showed a moderately positive correlation (r = 0.41) while minimum temperature showed a moderately negative correlation (r = -0.57). These results suggest that climatic conditions, especially during the rainy season and cooler months, play a crucial role in influencing the increase in fruit fly population. Phenology alone is insufficient as a predictor, but when considered in interaction with climatic variables especially temperature, it becomes a strong determinant. The study strongly recommends a longer-term data set and the inclusion of the fruit fly reproduction process and additional environmental variables to improve prediction accuracy.
Housing and Residential Preferences Among Large-Scale Miners - A Case Study of Staff of Newmont Ghana Gold Limited
(UENR, 2025-06) Mensah, S.O.
This study examined the housing and residential preferences of large-scale miners at Newmont Ghana Gold Limited in the Ahafo Region, with the aim of identifying their preferred housing types, assessing the factors that influence these preferences, and exploring the challenges they face in the rental housing market. The study adopted a mixed-methods descriptive design under the pragmatist paradigm. Data were collected through questionnaires and semi-structured interviews from a sample of 149 respondents. Quantitative data were analysed using descriptive statistics, correlation, and regression analysis in SPSS version 23.0, while qualitative data were subjected to thematic content analysis. The findings revealed a strong preference for modern housing units, with 49.7% favouring semi-detached houses and 28.9% preferring apartments, largely due to privacy, space, and family comfort. Income levels, household size, and proximity to the mine emerged as the most significant determinants of housing choice. The study also identified major rental challenges, with advance rent payments (mean = 3.61; t = 5.99, p < 0.001) and poor housing maintenance (mean = 3.30; t = 2.96, p = 0.004) as the most critical issues. The study concludes that miners’ housing choices reflect rational decision-making consistent with Price Theory and the Theory of Consumer Behaviour as they seek to maximise satisfaction within the constraints of affordability, household needs, and work-related demands. It is recommended that the government strengthen enforcement of the Rent Act to curb unlawful advance rent demands, mining companies expand employer-provided housing schemes, and landlords adopt flexible payment and maintenance systems to improve tenant satisfaction. The study contributes to the literature on housing in resource-driven economies and offers practical insights for policy makers, mining firms, and housing developers seeking to improve housing accessibility and quality in Ghana’s mining regions.
Occupational Hazards Associated with Municipal Solid Waste Collection in Sunyani
(UENR, 2025-05) Nimako, N.C.
The collection and disposal of municipal solid waste (MSW) pose major occupational dangers to waste collection employees. Therefore, the management of MSW has both public health and environmental concerns. Waste collectors are often exposed to a myriad of occupational hazards, which are frequently overlooked. The main objective of the study was to assess the occupational hazards associated with solid waste collection in the Sunyani Municipality in the Bono Region of Ghana. The cross-sectional survey was adopted to sample 159 waste collectors from Zoomlion Ghana Limited, Sunyani Municipal Assembly, Hand in Hand Waste Management Organization, City Waste Engineering, and Derrico Waste Management and Construction Limited. Also, 5 managers from the five waste management companies were sampled and interviewed. It was found that 57.9% of the waste collectors were usually involved in the collection of medical waste which poses health hazards to the waste collectors. About 81.8% of the waste collectors indicated that they have ever encountered hazardous materials whilst collecting waste. Generally, compliance with hazard control measures among waste collectors is low with only 17.6% of the waste collectors indicated that they usually use Personal Protective Equipment (PPE). Female workers, older workers and those with less years of experience use PPE more frequently than their peers (P-values<0.05). The average cost of medical treatment for employees was GHC750 per year. The study therefore, concludes that waste collectors are exposed to several hazards and health risks in the line of their duty due to low compliance with safety measures, including the use of PPEs. The treatment of occupational-related injuries puts some financial burden on waste collectors. The study recommends that waste management companies should intensify health education among waste collectors on the hazards associated with their work and how to avoid these hazards.
Integrating Geospatial Science and Technology in Geography Education : Perceptions and Opportunities in Senior High Schools Ghana
(UENR, 2025-11) Adongo, T.
This study establishes a baseline for integrating Geospatial Science and Technology (GST) in Ghanaian Senior High School Geography during the early rollout of the standards-based curriculum. Guided by TPACK/G-TPACK and TAM/UTAUT2, it examines current applications, teacher capacity, administrative support, student readiness and opportunities/constraints. A cross-sectional descriptive survey used stratified purposive and convenience sampling to collect questionnaires from students (n = 405), teachers (n = 25) and administrators (n = 15). Instrument quality was supported by expert review and internal consistency (Student Attitudes triad α = .79). Quantitative data were analysed in R (RStudio); open responses were thematically coded. GST use was limited and uneven (55.8% no prior exposure; 44.2% mainly via web mapping), yet awareness was high (74.3%) and support was strong (99%; 72.8% for immediate implementation). Teachers reported moderate familiarity with uneven preparation; administrators unanimously endorsed GST but reported few current initiatives. Inferential results reinforced the pattern: awareness differed by gender (small–to–moderate), attitudes did not differ by gender, and attitudes differed by region (medium)—implicating facilitating conditions (training, infrastructure, timetable/assessment alignment) rather than inherent preferences. A curriculumaligned, phased web-GIS pathway is recommended, pairing short, in-class PD with minimal ICT provision and leadership monitoring. Given the non-probability design, findings support analytical generalisation to comparable SHS contexts.
Modelling Road Traffic Crashes in the Sunyani Municipal using Gis
(UENR, 2025-02) Asare, S.K.
Road traffic accidents continue to be a major global public health and safety concern, requiring creative mitigation strategies. Using Sunyani Municipal as a case study, this study investigates how effective Geographic Information Systems (GIS) can be applied to model road traffic crashes to influence decision-making and reduce crash occurrence. This study identifies highrisk areas, assesses contributing factors, and suggests data-driven remedies by combining geographic analysis with collision data, road infrastructure characteristics, and traffic flow patterns. Road network layouts, traffic volume figures, and police crash reports from 2018 to 2022 were among the secondary data that were georeferenced and examined using GIS methods such as spatial autocorrelation, hotspot analysis, and Kriging. The findings showed clear spatial-temporal clusters of crashes, with a significant Moran's I index of 0.080 (z-score = 2.49, p-value = 0.006), confirming non-random clustering. Hotspot analysis (Getis-Ord Gi*) identified the Sunyani Technical University to Sunyani Senior High School stretch as the most critical hotspot with a 99% confidence level, alongside several other corridors (e.g., Estate Junction to Post Office) at a 95% confidence level. Key contributing factors with high respondent agreement included over-speeding and careless driving (18.67% each) and wrongful U-turns (14.67%). Socioeconomic elements that increased the crash risk included proximity to schools and markets. It is important to note that these findings are subject to uncertainties, primarily due to limitations in data completeness and potential geolocation inaccuracies inherent in the police-reported crash data. The study illustrates how GIS may be used to visualize risk patterns, which aids policymakers in prioritizing infrastructure improvements, implementing focused traffic laws, and streamlining emergency response routes. Installing traffic-calming measures in designated hotspots, improving street lighting, and incorporating realtime GIS monitoring systems are among the recommendations. This study emphasizes the importance of GIS as a tool for data-driven road safety management, providing urban governments facing comparable difficulties with scalable insights. The findings support the implementation of GIS-driven policies in Sunyani Municipal and similar places to promote safer road ecosystems and lower crash related injuries and deaths to materialize the country's UN goal by 2030.
Assessing the Effect of Illegal Mining and Food Security in Juaboso District of Ghana
(UENR, 2025-09) Armah, N.
The main objective of this study is to assess the socioeconomic factors that influence farmers’ participation in illegal mining, food security and the effect of illegal mining on land and water in the Juaboso District. The sample frame for the study were farmers both present farmers and those who have switched to illegal mining. The study made use of the mix method approach which had to do with both qualitative and quantitative research design. Overall, 120 respondents were interviewed for the study. Primary data was collected by means of structured questionnaires. Binary logistic model was used to isolate the factors that influence farmers’ participation in illegal mining. In addition, household food insecurity access score and dietary diversity was used to assess the respondents food security status. The study showed that illegal mining has severely degraded farmland and water resources while diverting labour from agriculture, leading to farm abandonment and rising conflicts. Further, Age, sex and household size were the main socioeconomic factors that drive farmers to switch from farming to mining:. Younger farmers, households with larger sizes and males were the main group of farmers that switched from farming to illegal mining. Analysis of the respondents food security status showed that farmers were more food insecure and miners (former farmers) enjoyed higher food security and dietary diversity through cash income from mining. It is thus recommended that government and local traditional rulers should enforce policies to protect our resources and create sustainable livelihood programs targeting youths and large households.
Assessment Of Nature-Based Solutions To Climate Variability and Adaptation Strategies Among Tomato Farmers in Techiman North Municupal Ghana
(UENR, 2025-11) Enchill, G.F.
The concept of nature-based solutions has emerged as a comprehensive strategy worldwide for addressing complex societal issues, such as climate change. This study explores the naturebased climate adaptation strategies adopted by tomato farmers in Techiman North Municipality of Ghana. Using a mixed-methods approach and a cross-sectional design, data were collected from 255 farmers through structured questionnaires, interviews, and field observations. Quantitative data were analysed using SPSS (version 26.0), applying descriptive statistics, onesample t-tests, and binary logistic regression, while qualitative data were analysed thematically. Findings showed that farmers widely use strategies such as restoring natural vegetation (33.6%), contour farming (32.4%), organic fertilizers (31.6%), agroforestry (30.5%), and intercropping (29.3%). Composting, buffer zones, and rainwater harvesting were rated the most effective nature-based adaptation strategies (mean > 3.08), while crop rotation was rated significantly less effective (M = 2.78, p = .015). Overall, 73.8% of farmers believed that their adaptation efforts were successful. Logistic regression revealed that the strategy for adoption was influenced by access to training (p = .040), perceived profitability (p = .039), traditional beliefs (p = .019), and policy constraints (p = .025). These results highlight the role of cultural norms, shared experiences, and institutional support in shaping farmer responses. The study concludes that smallholder farmers are not just responding to climate challenges but are proactively using nature-based solutions to improve resilience. Recommendations from the study include strengthening farmer training, promoting peer learning, enhancing policy support, improving land access, and expanding financial support for scaling up effective strategies.
The Relationship Between Diameter at Breast Height, Height, And Crown Volume of Teak Plantations in Ghana:Implications for Biomass Carbon Stock Estimation
(UENR, 2025-09) Azagbre, E.
Plantation forests are essential in mitigating climate change due to their high carbon sequestration and biomass accumulation potential. Tectona grandis (Teak) is a widely cultivated, fast-growing hardwood in Ghana, valued for its economic benefits, role in afforestation, biodiversity conservation, and contribution to national carbon accounting. However, accurate biomass and carbon estimation in Ghanaian Teak plantations remains limited by reliance on generalized allometric models developed under different ecological conditions, leading to potential errors in carbon assessments. This study aimed to improve understanding of allometric relationships between diameter at breast height (DBH), total height, and crown volume across Teak age classes (5, 10, 15, and 25 years). It also examined the influence of field-measured versus allometrically estimated heights on carbon stocks and evaluated Teak’s carbon sequestration potential. Data were collected in Form Ghana Limited plantations in Ghana using simple random sampling design, with measurements of DBH, height, and crown dimensions. Biomass and carbon stocks were derived using published allometric equations for Teak. Regression and one-sample T-tests was done to assessed the conformity to geometric and elastic similarity theories. Findings showed a strong DBH – height relationship but DBH – crown volume showed weak and inconsistent relationships with crown volume. highlighting the influence of competition and light availability on crown and DBH development. Carbon accumulation peaked at 15 years, followed by 25, 10, and 5 years. Field Measured height-based estimates were consistently higher than model-derived height values. The study recommends age-specific, locally calibrated allometric equations and the integration of direct crown measurements or remote sensing tools like LiDAR to improve carbon estimation precision.
Mitigating Hallucination in Large Models: A Modular Framework for Detection and Counterfactual Correction
(UENR, 2025-12) Nyantakyi, A.B.
Large Language Models (LLMs) demonstrate impressive fluency yet remain unreliable in safetycritical environments due to persistent hallucination; confidently generating factually incorrect or semantically not supported answers. This research proposes a modular mitigation framework integrating Hallucination Potential Minimization (HPM) with Self-Generated Counterfactual Training (SGCT) to improve factual consistency in generative outputs. A lightweight DistilBERT-based HPM classifier was trained as a binary factuality judge using benchmark datasets including FEVER and TruthfulQA, prioritising recall to ensure conservative hallucination detection. Building on this foundation, SGCT fine-tuned a GPT-2 generative model rather than more recent architectures due to its computational accessibility, reproducibility, and suitability for controlled experimentation under resource constraints. SGCT incorporates likelihood loss for factual responses, unlikelihood loss to penalize hallucinations, and a contrastive objective to separate factual versus hallucinated answers representations in an embedding space. Experimental results demonstrated measurable improvements following SGCT, with accuracy increasing from 0.556 to 0.614, recall from 0.705 to 0.890, precision from 0.532 to 0.548, and F1- score from 0.607 to 0.692. Threshold calibration further revealed flexible trade-offs between factuality and output strictness, enabling uncertain responses to be routed into a safe “abstain” category. The findings indicate that classifier-guided generation provides a practical strategy for enhancing reliability in LLM-based systems while maintaining computational efficiency. The proposed SGCT-HPM pipeline represents a reproducible and adaptable approach for hallucination mitigation, with potential applications in domains requiring verifiable AI-generated content.
Gamified Learning Applications for Children
(UENR, 2025-09) Adomako, O.S.
This research addresses the critical challenge of personalization in early childhood educational applications through the design, development, and empirical evaluation of QuestKids, a gamified learning application featuring a novel Dynamic Difficulty Adjustment (DDA) algorithm. While gamification has demonstrated potential for enhancing engagement in learning environments, most existing applications for young learners (ages 5-6) lack robust technical mechanisms for real-time, performance-driven adaptation, resulting in either frustration or boredom due to static content progression. The primary contribution of this work is the implementation and validation of a rulebased DDA algorithm that dynamically modulates question difficulty based on a rolling average of user performance, maintaining learners within an optimal "flow zone" of 65-85% success rate. Developed using the Flutter framework with a V-Model methodology, QuestKids integrates adaptive learning modules for mathematics and reading with a comprehensive gamification engine featuring badges, virtual currency, and progression systems. A comparative evaluation study was conducted with 85 children over six weeks, benchmarking QuestKids against two established educational applications (Duolingo Kids and Khan Academy Kids). Results demonstrated that QuestKids achieved significantly higher engagement metrics, with 56% longer average session duration and 78% daily return rate compared to benchmark applications. Learning outcomes showed a 30% higher knowledge gain and 85% retention rate after one week. Statistical analysis confirmed the superiority of the adaptive approach, with significant differences in both engagement (p < 0.01) and learning gains (F=6.34, p=0.002). The study concludes that a carefully engineered DDA algorithm, integrated within a gamified learning framework, can effectively enhance both engagement and educational outcomes in early childhood education. This research contributes to educational technology by providing a validated technical model for adaptive learning systems and offering practical insights for developers and educators seeking to create more personalized, effective digital learning tools.
Support Vector Regression for Forecasting Ghana’s Usd to Ghs Exchange Rate_A Comparative Study With Econometric and ML/QML Baselines.
(UENR, 2025-11) Anhwere, K.L.
Ghana continues to struggle with accurate exchange-rate forecasting because to structural breaks, nonlinear macroeconomic dynamics, and the shortcomings of conventional econometric models like ARIMA, ARIMAX, and VAR. These Bank of Ghana institutional workhorses are predicated on linearity, stationarity, and steady parameters—conditions that are increasingly broken in the fluctuating USD/GHS market. Studies that have already been conducted in Ghana frequently benchmark forecasting models inconsistently, lack unified assessment pipelines, and fail to evaluate operational usability or computational efficiency. The question of whether contemporary machine-learning approaches can produce much better and more trustworthy forecasts for decision-making is thus left open from a methodological and policy standpoint. This study investigates whether a kernel-based Support Vector Regression (SVR) model provides better operational robustness and prediction accuracy than the baseline models used by the Bank of Ghana. We employ a two-phase strategy using a monthly macroeconomic panel spanning 2014– 2023: (i) an ex-post backtest (train 2014–2022; test 2023) and (ii) an ex-ante operational forecast for January–December 2024 following model refitting on data through December 2023. Using accuracy metrics (MAE, RMSE, MSE, MAPE, R²), computational cost indicators (training time, inference time, peak memory), and uncertainty quantification (native intervals for econometric models, bootstrap intervals for ML/QML), all models, econometric, classical ML, and exploratory quantum ML, are assessed within a single, leakage-safe pipeline. According to the 2023 backtest results, SVR significantly outperforms ARIMA and VAR, although ARIMAX is still the most effective conventional comparator. While quantum variations show promise but inconsistent accuracy, classical machine learning methods (LSTM, XGBoost) offer moderate increases. A policy-ready Streamlit prototype that operationalizes model selection, scenario analysis, uncertainty reporting, and exportable outputs for institutional usage is shown in the study's conclusion.
A Comparative Study of Classical and Quantum Support Vector Machines for Distributed Denial of Service Attack Detection on Network Traffic Data
(UENR, 2025-12) Adomako, N.J.
The escalating frequency and sophistication of Distributed Denial of Service (DDoS) attacks pose a significant threat to the integrity and availability of modern digital infrastructures, particularly in developing countries like Ghana. Traditional Intrusion Detection Systems (IDS) which are reliant on static rule sets, struggle to detect zero-day and high-volume attacks, necessitating more adaptive and intelligent solutions. Classical machine learning models such as Support Vector Machines (SVMs) have demonstrated strong classification capabilities in network traffic analysis; however, they face limitations in scalability, latency, and efficiency when applied to highdimensional or encrypted data streams. This study presents a comparative analysis of classical SVM and Quantum Support Vector Machine (QSVM) models for Distributed Denial of Service attack detection using the publicly available Software Defined Network Distributed Denial of Service dataset. The classical Support Vector Machines wasimplemented using Scikit-learn’s Support Vector Classifier class with a Radial Basis Form kernel, while the Quantum Support Vector Machine (QSVM) leveraged Quantum Information Science Kit’s Quantum Support Vector Classifier (QSVC) class with a ZZFeatureMap and quantum kernel evaluated on simulated backends. Both models underwent identical preprocessing, including data cleaning, one-hot encoding, feature scaling, and dimensionality reduction via Principal Component Analysis (PCA). Experimental results revealed that the classical Support Vector Machine outperformed Quantum Support Vector Machine in terms of accuracy, precision, recall, and Receiver Operating Characteristic curve performance, achieving 95.7% accuracy compared to Quantum Support Vector Machine’s 82.5%. However, Quantum Support Vector Machine demonstrated potential scalability benefits and theoretical advantages in kernel expressivity. The study highlights the practical limitations of current quantum simulators and emphasizes the need for hardware advancements to fully realize Quantum Support Vector Machine’s promise in cybersecurity contexts. By bridging classical and quantum approaches, this research contributes empirical insights to the emerging field of Quantum Cybersecurity and offers guidance for deploying hybrid detection systems in real-world network environments. The findings are particularly relevant for cybersecurity practitioners, researchers, and policymakers seeking next-generation solutions to evolving cyber threats.
Modelling the Dynamics of Conjunctivities Along the Tano Basin
(2024-09) Ampofo, D.S.
Ghana’s Tano Basin continues to be a major public health hazard for conjunctivitis due to the introduction of adenovirus and herpes simplex virus into river bodies by activities of anthropogenic and zoogenic. The study objectives were to develope a mathematical model to explain the dynamics of conjunctivitis in the the Tano basin, the local analysis on the model was performed and behaviour of key parameters at different α values were investigated. In order to account for the effects of memory and non-locality, this work modelled the dynamics of conjunctivitis transmission using Caputo-Fabrizio differential equations. For numerical simulation, the Lagrange polynomial iterative scheme is applied. We examine the model at different α values, which span from 1 to 0.65, in order to evaluate the effect of memory degradation on the spread of disease. The findings demonstrate that the dynamics of the disease are considerably changed by lowering α values, with lower values (α < 0.80) displaying more oscillations and a lower peak incidence. It is also discovered that α affects the basic reproduction number R0, with R0 falling as alpha falls. According to the research, taking memory effects into consideration while analysing the spread of conjunctivitis can enhance model accuracy and guide sensible management measures.

Browse

Recent Submissions