http://ir.tum.ac.ke/handle/123456789/174 Contains PDF journal articles for this department Sat, 13 Jun 2026 18:33:24 GMT 2026-06-13T18:33:24Z http://ir.tum.ac.ke/handle/123456789/17605 Ensemble Network Intrusion Detection Model Based on Classification & Clustering for Dynamic Environment Muthama, Musyimi Samuel; Mwangi, Prof. Waweru; Calvin, Dr. Otieno - Anomaly detection is a critical issue in Network Intrusion Detection Systems (NIDSs). Most anomaly based NIDSs employ supervised algorithms, whose performances highly depend on attack-free training data. However, this kind of training data is difficult to obtain in real world network environment. Moreover, with changing network environment or services, patterns of normal traffic will be changed. This leads to high false positive rate of supervised NIDSs. Unsupervised outlier detection can overcome the drawbacks of supervised anomaly detection. Therefore, study apply one of the efficient data mining algorithms called ensemble network intrusion detection model based on classification & clustering. Without attack-free training data, ensemble clustering algorithm can detect outliers in datasets of network traffic. In this paper, study discuss model of anomaly-based network intrusion detection. In machine learning, a combination of classifiers, known as an ensemble classifier, often outperforms individual ones. While many ensemble approaches exist, it remains, however, a difficult task to find a suitable ensemble configuration for a particular dataset. This paper proposed method includes an ensemble feature selecting classifier, data mining classifier. The former consists of four classifiers using different sets of features and each of them employs a machine learning algorithm named - bagging-randomization -boosting and -stacking. The latter applies data mining technique to automatically extract computer users’ normal behavior from training network traffic data. The outputs of ensemble feature selecting classifier and data mining classifier are then fused together to get the final decision. The study proposes an ensemble-based that analysis of algorithm performance for intrusion detection. The method combines the output of four clustering methods to achieve an optimum selection. study then perform an extensive experimental evaluation of our proposed method using intrusion detection benchmark dataset, NSL-KDD. Thu, 01 Feb 2018 00:00:00 GMT http://ir.tum.ac.ke/handle/123456789/17605 2018-02-01T00:00:00Z http://ir.tum.ac.ke/handle/123456789/17604 EFFECT OF INVESTMENT PORTFOLIO CHOICE ON THE FINANCIAL PERFORMANCE OF INVESTMENT COMPANIES LISTED AT THE NAIROBI SECURITIES EXCHANGE Shukrani, Kenga Dominic; Ifire, Georgina Wanga; Yeya, Umulkulthum Musa; Banafa, Abdulkadir Ali Investment is a fundamental financial decision that both businesses and the general public ought to be aware of. Nonetheless, it is imperative to reminisce that every decision has consequences. The effect of investment portfolio choice on the financial performance of investment companies listed at the Nairobi Securities Exchange was investigated in this research study. Specifically, the study endeavored to investigate the effect of investment in bonds, investment in equities, and investment in real estate on the financial performance of investment companies listed at the Nairobi Securities Exchange. The modern portfolio theory, the efficient market hypothesis, the behavioral finance theory, the liquidity preference theory and financial intermediation theory informed this research study. Secondary data was used in this research. The study adopted a descriptive research design in analyzing the effects of the study variables. Several diagnostic and correlations tests were conducted before ultimately running the multiple linear regression model used in modeling the results of this study. The correlation results indicated a strong positive relationship between investment in bonds, equities and real estate with financial performance. Hypothesis testing at 5% level of significance established a significant effect on investment in bonds and investment in real estate, thus leading to the rejection of H01 and H03, while H02 was accepted. It was concluded that investment in bonds and real estate significantly affect the financial performance of investment companies listed at the Nairobi Securities Exchange. Thus, close monitoring and awareness creation to investors, governments and the general public around these variables is paramount for informed investment decision making. Fri, 24 Jun 2022 00:00:00 GMT http://ir.tum.ac.ke/handle/123456789/17604 2022-06-24T00:00:00Z http://ir.tum.ac.ke/handle/123456789/17602 A Hybrid and Adaptive Non-Local Means Wavelet based MRI Denoising Method with Bilateral Filter Enhancement Kagoiya, Kenneth; Mwangi, Elijah Magnetic Resonance Imaging is one of the most advanced and effective medical diagnosis methods ,however the raw image data is normally corrupted by random noise from the measurement process this reduces the accuracy and reliability of the results. Denoising methods are often used to increase the Signal-to-Noise Ratio (SNR) and improve image clarity .In this paper an adaptive Non-Local Means filter is developed in which bilateral filter is used to pre-enhance the images and then multi resolution wavelet domain is used to remove coefficients that contain more noise than signal. In the past different methods have been used to denoise MRI images but many have not taken into consideration the Rician nature of noise distribution therefore they have not been very effective .Adaptation in this case is based on frequency and spatial information obtained from the noisy image. Knowledge of level of noise is used in an optimization procedure to minimize a Rician based likelihood function and by use of square signal intensity bias is also discarded. The method is implemented in Matlab and MRI images with different level of artificial noise are denoised using the algorithm. Measures of performance values are PSNR, 37.12dB, MSE, 15.23, UQI, 0.985, SSIM, 0.894 , EPI,0.69 for a 10% noisy image. These and also visual inspection show that there is significant improvement from results obtained using stand alone methods such as Gaussian smoothing, Wiener filter, NLM filter ,bilateral filter and wavelet thresholding. General Terms Medical image processing, denoising, measures of quality, magnetization vector, adaptive multi resolution, total variation, wavelet coefficient thresholding ,Non-Local Means, frequency localization, proton spin density, discrete inverse Fourier transform, phase unwrapping, signal-dependent bias, local neighbourhood, Rician-based log-likelihood function. Sun, 01 Jan 2017 00:00:00 GMT http://ir.tum.ac.ke/handle/123456789/17602 2017-01-01T00:00:00Z http://ir.tum.ac.ke/handle/123456789/17582 INVESTIGATION ON SELECTED AREAS DEVIATING FROM THE SECOND LAW OF THERMODYNAMICS-CHALLENGES TO THE THERMODYNAMICS’ SECOND LAW Munialo, Patrick Wanyonyi; Enwemeradu, Christopher K.; Owuor, James; Cherop, Peter T.; Makokha, Seth Challenges to Thermodynamics’ Second law In this paper we undertake an investigation of the studies of the areas that show challenges to the second law of thermodynamics. This investigation has identified two areas causing a challenge to the second law of thermodynamics. The identified areas are Magneto Caloric Effect and Little Parks Effect DOI: https://doi.org/10.17605/OSF.IO/73R Sun, 01 May 2022 00:00:00 GMT http://ir.tum.ac.ke/handle/123456789/17582 2022-05-01T00:00:00Z