http://erepository.uonbi.ac.ke/handle/11295/82048 2026-05-18T20:39:55Z 2026-05-18T20:39:55Z Waweru, Salome W http://erepository.uonbi.ac.ke/handle/11295/161606 2022-11-02T10:48:40Z 2022-01-01T00:00:00Z

Improvement of Existing Intrusion Detection Systems Through Neural Networks Waweru, Salome W Recently, there is an increased rate of improved and unknown intrusions in the Intrusion detection system (IDS) field. In this study, artificial neural networks (ANN) model is proposed and integrated to develop the IDS execution on classification. Continual semi-supervised learning is the primary method conducted on artificial neural networks to create the model. Neural networks undergo pre-training with extensive benchmark datasets used for intrusion detection depending on how the user profiles have been created that have exact simulation of events and behaviors that are checked on the network. The evaluation domain is summarized from a unique set of features from the data set. The dataset are different due to the different user profiles. The ANN are trained for diverse attacks like Network infiltration, DDOS, Web attacks and many other arising attacks. The continual learning allows the model to learn latest advancements in deep learning and updates on recent anomalies detected. This improves the IDS to detect anomalies accurately. The main contribution of the proposed model to mitigate the gap identified is to assist in better classification of data, through the continual learning of neural networks, so that false positives that pass as normal data can be reduced in the system. It is demonstrated with experimental findings that the technique proposed can maintain a real time feedback to the anomaly.

2022-01-01T00:00:00Z Ogola, Jacob A http://erepository.uonbi.ac.ke/handle/11295/154619 2021-02-03T07:59:20Z 2020-01-01T00:00:00Z

Numerical Solutions of Fredholm Integral Equations of the Second Kind Research Report in Mathematics, Number 30, 2020 Ogola, Jacob A This thesis mainly focuses on the Mathematical and Numerical aspects of the Fredholm integral equations of the Second Kind. Due to its wide range of physical applications, we are going to deal with three types of equations namely: differential equations, integral equations and integro-differential equations. Some of the applications of integral equations are heat conducting radiation, elasticity, potential theory and electrostatics. Generally, we define an integral equation where an unknown function occurs under an integral sign. Also integral equations can be classified according to three different dichotomies: 1. Nature of the limits of integration 2. Placement of the unknown function 3. Nature of the known function After the classification of these integral equations we will have to investigate some analytical and numerical methods for solving the Fredholm integral equations of the Second Kind. Analytical methods include: degenerate kernel methods, Adomain decomposition methods and Successive approximation methods and Numerical Methods include: Degenerate kernel methods, Projection methods, Nystrom methods and Spectral methods. The main objective of the thesis is to study Fredholm integral equations of the Second Kind. In chapter 4 we have given the approximate methods (Spectral Methods) to solve these equations Using The Classical Orthogonal polynomials which is the main idea of this thesis where we apply the Spectral Approximation Methods for approximating the Fredholm integral equations of the Second Kind.

2020-01-01T00:00:00Z Camara, Mounirou http://erepository.uonbi.ac.ke/handle/11295/153668 2021-01-19T09:43:29Z 2020-01-01T00:00:00Z

Comparison Of Reverse-Transcription Polymerase Chain Reaction, Rapid Immunochromatographic Diagnostic Test And Fluorescent Antibody Test For Detecting Rabies Viruses Circulating In Mali Camara, Mounirou In Mali, a rabies reporting procedure is in place but is efficient only in the capital city where the Central Veterinary Laboratory (CVL), which is mandated to diagnose rabies, is located. This has led to an underestimation of the diagnosis of rabies including the genetic characterization of the virus in the country. Therefore, there is need to evaluate the diagnostic methods of rabies for subsequent characterization of circulating rabies virus in Mali. In this regard, the study assessed the suitability of the Rapid Immunochromatographic Diagnostic Test (RIDT), and Reverse-Transcription Polymerase Chain Reaction (RT-PCR) for the detection and characterization rabies viruses circulating in Mali in 2017. A total of 18 samples previous submitted to the CVL in Mali were analysed for rabies virus using the lateral flow device (BioNote, Inc., Seoul, Korea) and RT-PCR. RT-PCR positive samples were sequenced using Sanger sequencing method at Inqaba Biotec and subjected to phylogenetic analysis. In order to compare to two methods, Fluorescence Antibody Test (FAT) was used as the gold standard method. Out of the 18 samples, 16 were found to be positive for rabies virus on FAT. Out of these 16 positives, only 7 (43.8%) samples were positive for the virus on RIDT while 15 (93.8%) samples were positive for the virus on RT-PCR. All the sequences analysed by Blastn shared at least 93.5% nucleotide identity to the rabies nucleoprotein gene thereby confirming rabies infection of dogs in Mali. A phylogenetic analysis revealed that all the sequences belong to the Africa 2 lineage of which five to the sub-lineage H, four to the sublineage F and two to the sub-lineage G. The results of RT-PCR were comparable to those of FAT. However, the positivity detection rate for RIDT was low as compared FAT. The genetic characterization of the virus confirmed previous findings of the circulation of the sub-lineages H, F and G belonging to the Africa 2 lineage in Mali. In conclusion, the RT-PCR could be used together with FAT for the detection and genetic characterization of rabies virus circulating in Mali. Further studies using large number of samples are required to validate the suitability of the new RIDT for the diagnosis of rabies in Mali.

2020-01-01T00:00:00Z Menjeri, Obed M http://erepository.uonbi.ac.ke/handle/11295/104635 2019-01-14T17:18:17Z 2018-01-01T00:00:00Z

Strategic Asset Allocation of Pension Funds; An Application of Markowitz Portfolio Theory Menjeri, Obed M The management of pension funds is a very sensitive and important aspect that has a bearing on the quality of life given by retirees when they come to the end of their useful work life. Therefore their risk management is very crucial. The financial products n which the pension assets are invested in have different levels of risk that investment managers are loo of. Their aim is therefore to maximize the returns for their members while taking the minimum possible risk with their resources. Markowitz provided a solution to this problem through the mean-variance model, which has been critically analyzed and used to various investment portfolios. This study was therefore aimed at finding an optimal way of allocating the pension funds keeping in mind their risk characteristics. The results from the analysis done showed tremendous improvement in terms of efficiency in the allocation of the pension assets to various investment opportunities. The optimal restricted portfolio gave us a return of 9.47% while the unrestricted one gave us 13.45%. this came with a standard deviation of 10.45% and 13.74% respectively. Therefore, the investor can be able to invest 62% and 60% of the restricted and unrestricted portfolios respectively in the risky portfolio and 38% and 40% respectively in the risk free asset. From the data used for this study, this will give the best returns. This thesis also explores the improvements that can be made to index funds by removing the link between pricing errors and portfolio weights and compared their performance with that of actively managed funds. Index funds today tend to overweight over valued companies, leading to serious performance lags especially during pricing bubbles. The Markowitz optimization formulation is used in combination with the fundamental metrics and the weightings from the solutions to the mean variance optimizations were used to calculate the expected returns. The Sharpe ratio for the S&P 500 was 0.07 and that of the Mean Variance Optimization was 0.57. The results clearly demonstrate that the MVO outperforms the S&P 500. By using fundamental metrics such as P/E ratio, sales, book value and dividends to evaluate the size of a company rather than the traditional market capitalization, significant improvements can be made to the value of a portfolio.

2018-01-01T00:00:00Z