http://hdl.handle.net/20.500.12283/351 Wed, 15 Apr 2026 01:31:13 GMT 2026-04-15T01:31:13Z http://hdl.handle.net/20.500.12283/4661 Comparative assessment of random forest and support vector machine in gold vein prediction Nampeela, Saidah Sanyu Accurate forecasting of ore veins is essential in mining due to the intrinsic variability in vein geometry, grade, and orientation, often complicated by the nugget effect which challenges conventional resource estimation. Traditional geostatistical methods like kriging frequently struggle with the complex geometries and skewed grade distributions typical of vein deposits. This research evaluates the performance of machine learning algorithms, specifically Random Forest (RF) and Support Vector Machines (SVM), for predicting the presence and continuity of gold veins in the Tiira Gold Fields, Uganda. Utilizing geological survey data (gold concentration) combined with geophysical measurements (magnetic intensity, resistivity, chargeability) and geospatial coordinates, predictive models were developed following rigorous data integration, cleaning (using Inverse Distance Weighting for interpolation), and preprocessing. A comparative assessment of RF and SVM was conducted for both binary classification (vein presence > 10 ppm) and regression (gold concentration prediction). Results indicated strong performance for both models in classification, with RF achieving a higher Area Under the Curve (AUC) (0.953) but SVM showing better precision (0.909) on test data. For regression, SVM demonstrated superior performance in predicting gold concentration (Test R² = 0.718) compared to RF (Test R² = 0.617). Sensitivity analysis identified chargeability and resistivity as the most influential predictors. This study provides a robust framework for applying machine learning in mineral exploration within complex geological settings, aiding informed decision-making and potentially reducing unnecessary excavation, thereby supporting sustainable resource management. Dissertation Wed, 01 Jan 2025 00:00:00 GMT http://hdl.handle.net/20.500.12283/4661 2025-01-01T00:00:00Z http://hdl.handle.net/20.500.12283/4630 Investigating the effectiveness of ammonium thiosulfate in gold recovery. Banyahana, Jude With the increasing environmental pollution and health risks in communities within and around gold mining areas in Tiira, Uganda, there is a growing need to find safer gold extraction methods. The government of Uganda, along with NGOs like the National Environmental Management Authority (NEMA) and the World Health Organization (WHO), has identified cyanide and mercury as major pollutants in artisanal and small-scale gold mining (ASGM). In Tiira, most small-scale miners rely on mercury amalgamation, which releases toxic mercury into the air, water, and soil. This exposure has severe health effects, including damage to the nervous system, kidneys, liver, and immune system. Additionally, mercury use in gold recovery results in impurities in the final gold product. This project investigates the effectiveness of ammonium thiosulfate (ATS) in gold recovery as a safer alternative to traditional methods. The study objectives are: to characterize the gold sample; to determine gold leaching efficiency of ammonium thiosulfate under varying conditions; and to optimize the operational parameters for maximum recovery. Chapter One introduces the study, presenting the background of the problem, clear research objectives, the scope of the study, and the justification for exploring ATS as an alternative gold recovery method. Chapter Two consists of the literature review, which examines previous research on gold extraction methods, comparing the efficiency, environmental impact, and feasibility of ATS leaching versus mercury amalgamation and cyanide leaching in small-scale mining. The chapter also highlights case studies where thiosulfate leaching has been successfully applied. Chapter Three covers the methodology used in the research. This involved collecting and characterizing a gold ore sample from Tiira, conducting laboratory tests under varying pH, temperature, and ATS concentrations, and analyzing the effectiveness of ATS using both qualitative and quantitative techniques. Chapter Four presents the results and discussions of all tests and experiments carried out. It includes a detailed analysis of the gold content in the sample, recovery efficiency under different conditions, and the optimization of parameters for improved leaching performance. The conclusion section summarizes the key findings, highlights the limitations and challenges encountered during the research, and provides recommendations for improving the use of ATS in small-scale mining. The references section cites various articles, journals, and reports that contributed to this study. Dissertation Wed, 01 Jan 2025 00:00:00 GMT http://hdl.handle.net/20.500.12283/4630 2025-01-01T00:00:00Z http://hdl.handle.net/20.500.12283/4627 Redesign and fabrication of a sluice box for improved portability and gold recovery Kumiki, Radwan Said The project addressed the critical inefficiencies in existing sluice box designs used by artisanal gold miners in the Tiira gold mines, Uganda. These inefficiencies included difficulties in feed and water delivery, suboptimal inclination angles and the use of materials that contributed to high mercury contamination, posing significant threats to both environmental health and miner wellbeing. To overcome these, the project aimed to redesign and fabricate a portable and efficient sluice box for artisanal gold miners in Tiira, Uganda, addressing inefficiencies in existing designs such as poor feed delivery, suboptimal inclination angles, and high mercury contamination. The research focused on four specific objectives which included determination of optimal sluicing parameters for improved recovery efficiency, development of a portable and efficient sluice box design, fabricating and testing the new design to evaluate its performance then lastly performing financial analysis to which determined the economic viability of the project. Through field surveys and laboratory analysis of ore samples, key sluicing parameters including slope angle, water flow rate, and feed rate were optimized using Response Surface Methodology (RSM) and Central Composite Design (CCD). The redesigned sluice box achieved a 20.1% increase in gold recovery (from 67.75% to 87.8%) by implementing an optimal slope angle of 13.5°, a sluice length of 1.3 m, and a width of 0.6 m. Sensitivity analysis revealed that slope angle contributed 40.65% to recovery efficiency, followed by velocity (22.76%) and flow rate (20.33%). Constructed from mild steel for durability and portability, the final prototype weighed 12.88 kg and featured foldable components for ease of transport. Financial analysis confirmed viability, with a 2.8-year payback period and a profitability index of 2.944. The project demonstrates that optimized sluice box designs can significantly enhance gold recovery while reducing environmental and health risks associated with mercury use in artisanal mining. Future work should focus on material/riffle optimization for better performance and cost. Essential training for miners on optimal techniques and sustainable methods is crucial. Subsequent efforts must address scaling up the design for mass production and wider adoption. Dissertation Wed, 01 Jan 2025 00:00:00 GMT http://hdl.handle.net/20.500.12283/4627 2025-01-01T00:00:00Z http://hdl.handle.net/20.500.12283/4626 Design and fabrication of a manually operated sample splitter for improved portability and cost-effectiveness to the local miners in Tiira Okite, Christopher Design and fabrication of a manually operated sample splitter for improved portability and cost-effectiveness to the local miners in Tiira Dissertation Wed, 01 Jan 2025 00:00:00 GMT http://hdl.handle.net/20.500.12283/4626 2025-01-01T00:00:00Z