Master Theses and Dissertations (MST)

Enhanced hybrid machine learning model for detecting denial of service attacks in internet of things networks.

2025-01-01T00:00:00Z

Enhanced hybrid machine learning model for detecting denial of service attacks in internet of things networks. Ngunyi, Beatrice Njeri Connecting devices across homes, healthcare, agriculture, transportation, and businesses through the Internet of Things (iot) has become a critical part of modern life. Nevertheless, the universal interconnection of iot devices introduces significant vulnerabilities, making them susceptible to cyber threats such as Denial-of-Service (dos). These attacks deny legitimate users access to services and cause financial losses due to Service Level Agreement (SLA) violations. Traditional approaches have proven to be insufficient in handling large-scale, diversified, and complex iot environments. This study developed an enhanced hybrid machine-learning model that integrates Isolation Forest (unsupervised anomaly detection) with Random Forest (supervised classification) to improve the detection of dos attacks in an iot network. Benchmark datasets, including NSL-KDD and CICIDS017, were obtained from the Kaggle open-data repository, where the complete datasets were collected using Octoparse web-scraping software. Octoparse enabled the automated extraction of the entire dataset without modification, ensuring that all available records were included in model training and evaluation. Experimental results demonstrated that the hybrid model achieved superior performance – Accuracy (97.8%), a Precision of 98.2%, Recall (96.8%), F1-score (97.5%), and ROC-AUC (98.1%) consistently outperforming standalone models. The ROC curve analysis confirmed the hybrid model’s discriminative strength, showing clear separation between normal and attack traffic. In addition, scalability tests revealed that the model scales effectively to large datasets, with projections up to 5 million records. These findings highlight both the consistency and superiority of the hybrid model compared to traditional ML- based IDS solutions. Academically, the research fills a gap by emphasizing hybrid ML approaches specifically tailored for dos detection in iot environments and practically delivering a robust, scalable, and real-time security solution for industries and organizations seeking to secure large- scale iot infrastructures. A thesis submitted to the Department of Computer Science and IT in the School of Computing and Mathematics in partial fulfillment of the requirements for the award of Degree of master of cyber security of the Co-operative University of Kenya.

Second hand car price prediction model in Nairobi.

2025-01-01T00:00:00Z

Second hand car price prediction model in Nairobi. Onyiego, Brian Atandi. The second-hand car market in Nairobi continues to grow rapidly, creating a need for accurate and transparent price prediction methods. Traditional valuation approaches rely heavily on subjective judgement, leading to inconsistent and unreliable pricing. This study aimed to develop a data-driven machine learning model capable of predicting second-hand car prices using structured vehicle characteristics such as year of manufacture, mileage, engine capacity, brand, and model. The population consisted of all vehicles listed on the SBT Japan online platform in Kenya. A total of 29,000 records were collected through web scraping, and after cleaning and preprocessing, 20,775 records were retained for analysis and modelling. Feature analysis showed that model, brand, engine capacity, year of manufacture, and mileage were the most influential predictors of price. Two ensemble learning models, Random Forest and Extreme Gradient Boosting, were developed and evaluated. The Extreme Gradient Boosting model achieved the highest accuracy, with a mean absolute error of 95,696.60 Kenyan shillings, a root mean square error of 190,939.99 Kenyan shillings, and a coefficient of determination of 0.99379, which represents a substantial improvement over the baseline error of 1,839,811.92 Kenyan shillings. The study concludes that machine learning provides a reliable, consistent, and highly accurate approach for predicting second-hand car prices in Nairobi, offering practical value to car dealers, financial institutions, insurers, online marketplaces, and potential buyers seeking transparent and data-driven pricing. A research project submitted to the Department of Computer Science in the School of Computing and Informatics in partial fulfillment of the requirements for the award of the master of Science in Data Science of the Co-operative University of Kenya.

A hybrid security model for data protection in public cloud computing.

2025-01-01T00:00:00Z

A hybrid security model for data protection in public cloud computing. Orucho, Daniel Okari This thesis sought to address the following objectives: assess security threats and attacks on data in public cloud computing, assess existing techniques for securing data in public cloud environments, develop a hybrid algorithm to enhance data security in public cloud computing, and evaluated the effectiveness of the proposed hybrid algorithm. This study adopted a positivist research paradigm and employed both descriptive and data science methodologies. Secondary data was collected from peer-reviewed journal articles, conference proceedings, and books, while high-quality images were sourced from the University of Southern California Signal and Image Processing Institute database for algorithm test simulations. Encryption was applied to numerical data before embedding it into cover images, and data analysis was conducted using content analysis, gap analysis, visual quality assessment, statistical evaluation, and comparisons with baseline algorithms. Simulations were performed using MATLAB R2021a on six color images. The study identified various threats to data in public cloud computing, including insecure APIs, legal and compliance issues, data corruption, code injection, insider threats, mobile malware, phishing and social engineering, denial of service, man-in-the-middle attacks, SQL injection, and cross-site scripting. These threats can be mitigated through robust security frameworks such as input validation, encryption, access controls, and continuous monitoring, along with organizational strategies like employee training, legal compliance audits, and regular vulnerability assessments. Techniques for securing data in public cloud environments include cryptographic algorithms like Blowfish, Twofish, RSA, digital signatures, homomorphic encryption, authentication, and data hiding methods. The proposed hybrid algorithm integrated Least Significant Bit substitution with Paillier Homomorphic Encryption and was evaluated using Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR), entropy, and histogram analysis. Results showed low MSE values (0.000023–0.00012), high PSNR values (87.3–94.57), and entropy values ranging from 6.4207 to 7.5937, indicating minimal distortion, high reconstruction quality, and strong data complexity. The algorithm maintained high visual and statistical fidelity post-embedding, with perfect correlation and stable entropy values, while minor chi-square fluctuations suggested localized changes without compromising imperceptibility or robustness. Overall, the hybrid algorithm proved effective for secure data hiding in public cloud computing by preserving image quality and ensuring statistical integrity. A thesis submitted to the department of computer science and information technology in the school of computing and Mathematics in partial fulfillment of the requirements for The award of the degree of master of science in cyber Security of the co-operative university of Kenya.

Leveraging machine learning models to predict HIV/AIDS treatment interruption in art patients in Machakos county,Kenya.

2025-01-01T00:00:00Z

Leveraging machine learning models to predict HIV/AIDS treatment interruption in art patients in Machakos county,Kenya. Odundo, Clifford Omondi HIV/AIDS remains a major global health challenge, with Sub-Saharan Africa carrying the highest burden. In Kenya, where adult prevalence is 4.3%, treatment interruption (IIT) continues to undermine antiretroviral therapy (ART) outcomes. This study aimed at applying machine learning (ML) to identify predictors of IIT and guide interventions in Machakos County, where prevalence is 3.3% and relies on manual appointment management of patients, physical tracing and phone tracing of patients. A retrospective cross-sectional study used secondary data from KenyaEMR covering 14,339 adults on ART between 2020 and 2024. Data preprocessing included cleaning, anonymization, imputation, encoding, LASSO feature selection, and SMOTE oversampling. Descriptive statistics and chi-square tests assessed associations, while Random Forest (RF), XGBoost, and Support Vector Machine (SVM) models were trained and validated to predict IIT. Overall, 910 patients (6%) experienced IIT. Risk was highest among adolescents and young adults (15–24 years), single individuals, urban residents, patients with viral load ≥1000 cps, those on ART <12 months, TB co-infected, and non-DTG regimen users. Poor adherence, unstable status, lack of phone ownership, and shorter refill durations also predicted IIT. Non-significant factors included sex, CD4 count, counseling, and clinic workload. Among models, RF achieved the best performance upon evaluation, (recall 0.97, precision 0.87, F1 0.92, AUROC 0.96, accuracy 0.91), outperforming XGBoost and SVM. IIT in Machakos County is shaped by demographic, clinical, socioeconomic, and health system factors. Random Forest showed the best predictive capacity, highlighting the value of ML for early identification of at-risk patients.Strategies should include DTG scale-up, early retention support, multi-month dispensing, and digital health interventions. Integrating predictive analytics into EMRs can strengthen HIV program outcomes. A project submitted to the department of computer science & Information technology in the school of computing and Mathematics in partial fulfillment of the requirements for the Award of the degree of master of science in data science of the Co-operative University of Kenya.