SU+ Digital Repository

SU+ is an online repository for the preservation and promotion of assorted digital content at Strathmore University

Off-Campus Access to restriced resources (including the ExamsBank) now requires registration using an @strathmore.edu email address

Authentication is NOT required for On-Campus Access to content

Photo by @inspiredimages

Communities in DSpace

Select a community to browse its collections.

Now showing 1 - 5 of 8

Conferences / Workshops / Seminars + 474
Documents and Proceedings of Conferences, Seminars, Workshops (and more) held at Strathmore University
Digital Archives 198
Assorted collections of resources covering various subject themes contributed by Faculty and Library Staff
ExamsBank 1949
Past examinations grouped according to the Programme of Study
Reports / Policies + 8
Public reports and policy documents
Research / Researchers / Publications 4335
Researcher Profiles / Conference presentations / Published research articles / Faculty and Corporate research outputs

Recent Submissions

A Chatbot for personalized financial advisory and investment planning in Kenya
(Strathmore University, 2025) Bukaya, S.
Financial illiteracy in Kenya remains a significant barrier to economic participation, limiting individuals' access to investment opportunities, insurance products, stock trading, retirement planning, and effective savings strategies. This research addressed the critical gap in affordable, transparent, and customized financial advisory services by developing an AI-powered chatbot tailored specifically for the Kenyan market. The study employed a mixed-methods approach combining exploratory, design-based, and experimental research methodologies to create and evaluate the chatbot system. Data was collected from 1,000 survey respondents representing diverse demographic and economic segments across Kenya, revealing key financial challenges and technology preferences. The chatbot architecture integrated natural language processing using BERT-based models, real-time market data from the Nairobi Securities Exchange and Central Bank of Kenya, and personalized recommendation algorithms. The system achieved 91.76% accuracy in intent recognition and 90.23% in parameter extraction across diverse financial domains. Notable innovations included bilingual support for English and Swahili with code-switching capability, mobile money integration, and culturally adapted financial advice. Performance evaluations demonstrated sub-50ms response times for 95% of queries, making the system accessible even on low-bandwidth connections. The chatbot successfully democratized access to professional-quality financial advisory services, particularly benefiting underserved populations who previously lacked such resources. This research contributes significantly to financial inclusion efforts in Kenya by demonstrating that AI-powered systems can effectively bridge the financial literacy gap when properly contextualized for local markets. The findings have important implications for developing similar solutions in other emerging economies facing comparable financial inclusion challenges.
An Ensemble machine learning model for drought prediction in Kilifi County, Kenya
(Strathmore University, 2025) Bosire, V. K.
Drought prediction is critical for mitigating its adverse effects on agriculture, water resources, and livelihoods. This study developed a stacking ensemble learning model for multi-class drought prediction using machine learning techniques, including Random Forest, Support Vector Machine (SVM), XGBoost, and LightGBM. The CRISP-DM methodology guided the research, ensuring a systematic approach to data collection, pre-processing, model training, evaluation, and interpretation. Each model was assessed based on accuracy, recall, precision, and F1-score to determine its effectiveness in classifying drought severity levels. Among the base models, Random Forest achieved the highest accuracy (62.79%) and recall (62.79%), demonstrating its ability to capture complex patterns in the data while maintaining a reasonable balance between precision and recall. In contrast, SVM performed the weakest, with an accuracy of 25.58% and an F1-score of 29.91%, indicating its limited effectiveness in handling the dataset's complexity and class imbalance. The XGBoost and LightGBM models showed moderate performance, with both achieving an accuracy of 55.81% and recall of 55.81%, but their high precision (90.93% for XGBoost and 83.97% for LightGBM) suggests they are effective in minimizing false positives, though they struggle to balance precision and recall effectively. The stacking ensemble model, which integrated these classifiers with Logistic Regression as the meta-model, yielded an accuracy of 65.12% and an F1-score of 62.68%, outperforming all individual base models. To further enhance model performance, the study recommends incorporating additional environmental variables, such as land surface temperature (LST), Standardized Precipitation Evapotranspiration Index (SPEI) and the Palmer Drought Severity Index (PDSI) to significantly improve drought prediction accuracy. Future research should explore deep learning approaches, hybrid models combining physical hydrological and machine learning methods, and explainability techniques like SHAP values or LIME to strengthen trust in AI-driven drought forecasting systems. Keywords: Drought Severity, Drought Prediction, Stacking, Ensemble Learning, Model Evaluation
Privacy-preserving machine learning tool for mitigating data leakage in microservices architectures
(Strathmore University, 2025) Sikolia, C.
In the era of distributed systems and microservices architectures, the risk of data leakage, particularly of personally identifiable information (PII), has become a critical concern. Federated Learning (FL) emerges as a promising solution by enabling collaborative model training across decentralized data sources without the need to transfer raw data to a central server, thereby preserving user privacy. This study implemented a privacy-preserving federated learning utilizing Flower frammework, an open-source FL platform, in conjunction with TensorFlow for model development. The Federated Averaging (FedAvg) algorithm was employed as the core aggregation strategy to combine model updates from multiple clients. A hybrid deep learning model was designed to optimize learning across the distributed network, ensuring both robustness and scalability. Over the course of five training rounds, two clients participated in each round, contributing locally trained model weights to a central aggregator. The performance of the federated model was rigorously evaluated using standard machine learning metrics: accuracy, loss, F1-score, Area Under the ROC Curve (AUC), and precision. The results demonstrated progressive improvement in model performance, with accuracy increasing from 76.0% in round 1 to 87.7% in round 5, and AUC improving from 0.88 to 0.93, indicating enhanced classification capability over time. Similarly, F1-score and precision showed consistent growth, signifying improved balance between precision and recall and reduced false positives. The distributed training also showcased a decreasing loss trend, dropping from 1.018 to 0.891 across the rounds, reflecting better model convergence. Importantly, this study illustrated the practical viability of federated learning for privacy-centric applications, showing that high-performance machine learning models can be achieved without compromising data privacy. Future work can focus on scaling this approach to a larger number of clients, integrating differential privacy and secure aggregation techniques to further strengthen privacy guarantees, and comparing the hybrid model with alternative architectures to enhance model generalizability across heterogeneous data sources. Keywords: Federated Learning, Privacy Preservation, Flower Framework, TensorFlow, Federated Averaging (FedAvg), Hybrid Deep Learning Model, Machine Learning Metrics, Aggregation
Flash floods prediction model: a case of Nairobi
(Strathmore University, 2025) Bico, S.
This study developed a flash flood prediction model for Nairobi, focusing on improving the city's resilience and emergency preparedness and response to flash floods. The study employed quantitative and experimental research design. A machine-learning model was built to predict flash flood occurrences using historical rainfall, soil moisture, and meteorological, hydrological, and topographical data. The key variables identified to influence flash flood occurrence were rainfall, soil moisture content, river discharge, and erosion degree, with rainfall showing the highest correlation (61%) to flash floods. Among various machine learning models, the Random Forest model outperformed others with an accuracy of 93.33%, recall of 90.47%, and an F1 score of 0.90, making it the most reliable predictor. Other models, such as KNN, Logistic Regression, SVM, and ANN, also showed impressive performance. The developed model can potentially improve flood prediction which can lead to reduction on damages, enhance Nairobi's resilience to flash floods, and providing a reference for other urban areas facing similar climate challenges. It was recommended that Nairobi and similar metropolitan areas should invest in enhancing their drainage infrastructure to complement the predictive model's capabilities. Integrating this model into city planning, emergency response systems, and early warning systems could help in mitigating the risks posed by flash floods. Additionally, policymakers should prioritize land use planning and environmental conservation to address the key drivers of flash floods, such as soil erosion and improper land use.
An Assessment of factors affecting the implementation of the Kenya Mental Health Policy 2015-2030
(Strathmore University, 2025) Wanjiku, S. M.
Mental health is a critical public health concern in Kenya, yet the effective implementation of the Kenya Mental Health Policy (2015–2030) faces significant challenges. Despite the policy's goal of ensuring accessible, affordable, and high-quality mental health services, systemic barriers such as underfunding, workforce shortages, stigma, and weak governance structures continue to hinder progress. This study assessed the factors affecting the implementation of the Kenya Mental Health Policy, with an appraisal of the policy, and lessons from international best practices. A qualitative research approach was employed, with data collected through interviews, document analysis, and expert consultations involving key stakeholders, including policymakers, mental health practitioners, and advocacy groups. Systems Theory was applied to examine how governance structures, funding mechanisms, and service delivery models interacted within Kenya’s mental health system. Policy Diffusion Theory provided insights into how global and regional trends influenced policy adoption and implementation. The findings revealed critical gaps, including poor intergovernmental coordination, inadequate data systems, and insufficient attention to adolescent mental health. Best practices from countries such as Ethiopia, South Africa, and Sweden were identified as potential models for strengthening Kenya’s policy effectiveness. By analyzing these factors, the study provided evidence-based recommendations to enhance governance, financing, and service delivery frameworks for the successful implementation of the Kenya Mental Health Policy (2015–2030).