https://cris.library.msu.ac.zw//handle/11408/325 Thu, 16 Apr 2026 11:00:22 GMT 2026-04-16T11:00:22Z https://cris.library.msu.ac.zw//handle/11408/3994 Title: Relating mathematics to machine learning through algorithm development for development for big data analysis Authors: Chirisa, Diamond Takudzwa Abstract: Data has increased at an exponential rate and has outpaced our capability to analyze it. However, new ways of data analysis, which thrive in big data such as Machine Learning (ML) have emerged. This study explores Machine Learning by creating a Machine Learning algorithm based on Support Vectors. This was done by converting mathematical formulations into a computer algorithm that was then used for data classification. The algorithm was evaluated and compared to other algorithms. The results of the evaluation show that the algorithm was accurate at binary classification. Comparisons to other algorithms using both the iris and breast cancer datasets show that algorithms based on Support Vectors are generally more accurate at data classification. This means that the approach that was used in this study can be used in businesses to determine whether a person will return loan or not or whether a particular student can finish a degree program or not based on past data. The study also indicated that Support Vector Machines algorithm training require more computing power as data gets bigger. Hence, it suggested use of high performance computing for big data analysis. Sun, 01 Jan 2017 00:00:00 GMT https://cris.library.msu.ac.zw//handle/11408/3994 2017-01-01T00:00:00Z Chirisa, Diamond Takudzwa https://cris.library.msu.ac.zw//handle/11408/3898 Title: Heat mass transfer with chemical reaction of an exponnetially stretching stenosed artery Authors: Majiri, Thandiwe Abstract: The laminar boundary layer flow in an exponentially stretching stenosed artery immersed in viscous and incompressible blood is investigated along with the effect of a chemical reaction. Many researchers have used mathematical modelling to predict blood flow through stenosed arteries. Pressure, shear stress and velocity are the parameters that have been analysed in the past. Not a lot of work has been done to highlight the effect of heat transfer on an exponentially stretching stenosed artery with a chemical reaction effect. In this research, stenosis is defined as a condition whereby arteries abnormally narrow. The ability to describe the flow of blood through a stenosed artery provides the possibility of diagnosing the related diseases even before they become clinically relevant. The governing partial differential boundary layer equations in Cartesian Co-ordinate form are first transformed into ordinary differential equations which are then solved by the Runge-Kutta and Shooting methods using the Matlab software package. Potential improvements to previous models have been made by the inco-operation of the effect of exponential stretching of the stenosed artery as blood flows and also by including the effect of a chemical reaction to blood flow. The researcher has come to conclude that the dimensionless temperature field depends on thermal diffusivity (α), the heat component (η) and the dimensionless (U0.D0)/ Re. U0 is the stretching velocity, D0 represents the length of the stenosed portion and Re is the Reynolds’ number. Also, chemical reactions mostly caused by foreign substances to the body, generally lower the flow of blood in arteries. The effects of asymmetric stenosis in a realm of the arterial plaque may be useful for early detection of cardiovascular diseases. Hence the researcher recommends that people take low fat and cholesterol diets as these highly lead to stenotic conditions. Stenotic conditions lead to diseases such as Athesclerosis, Coronary heart disease and high blood pressure. Sat, 01 Jun 2013 00:00:00 GMT https://cris.library.msu.ac.zw//handle/11408/3898 2013-06-01T00:00:00Z Majiri, Thandiwe https://cris.library.msu.ac.zw//handle/11408/3856 Title: A mathematical model of HIV/AIDS population dynamics with treatment failure and treatment dropouts in the era of universal test and treat approach Authors: Nhendo, Calvin Abstract: Antiretroviral therapy is currently the major intervention against HIV infection.However, with increased access to treatment through the universal test and treat approach, potential barriers to the overall success of this strategy such as treatment dropouts and treatment failure arise. We constructed a deterministic mathematical model of HIV/AIDS to study the possible effects of treatment failure and treatment dropouts on the population dynamics of the infection. The model incorporated a universal test and treat scenario and a separate sub population of treatment dropouts. The disease free and endemic equilibria is computed and the basic reproduction number R0 of the model, is determined using the next generation matrix method. Numerical simulations are presented to investigate the effect of treatment failure and treatment dropouts on the dynamics of the model and on the R0. From the expression of R0 it is shown that the treatment dropout class contributes to the overall model reproduction number. Results of the numerical simulations show that an increase in treatment dropouts leads to an increased transimission of the HIV infection in a population. Also, the results indicate that even in the absence of treatment dropouts and treatment failure the basic reproduction number remains above unit, highlighting the need for several control measures to end the epidemic. Treatment failure is shown to increase the maximum size of the AIDS class. The results from this study demonstrate the need to focus on increasing efforts of reducing treatment dropouts in combination with other intervention strategies, through monitoring adherence and identifying and enrolling back to antiretroviral therapy(ART) of treatment dropouts. Also there is need to improve on early diagnosis of treatment failure such that those on treatment do not progress to AIDS before they are put on second or third line ART. Tue, 01 Jan 2019 00:00:00 GMT https://cris.library.msu.ac.zw//handle/11408/3856 2019-01-01T00:00:00Z Nhendo, Calvin https://cris.library.msu.ac.zw//handle/11408/3714 Title: Morphological characterization of autochthonous sheep breeds at Matopos Research Station : a markov chain monte carlo algorithm approach. Authors: Chipindu, Lovemore Abstract: Failure to characterize autochthonous sheep breeds may result to the extinction of the important natural resource which can easily adapt to the local variation in climate change. Environmental factors and some human practices especially artificial insemination is leading to the evolving of different pedigrees. Price determination is an important aspect in designing pricing models, but what really contributes to the weight of autochthonous sheep breed remains as the major question which needs to be addressed. By merely looking at a sheep is it possible to deduce its body weight another question of interest arises again. This project addressed several issues associated with the characterization of autochthonous sheep by establishing the relationship between morphological traits and the body weight of an animal. The relationship was further addressed through the application of several methods namely the principal component analysis, generalized linear models, WinBugs model programming and the Markov Chain Monte Carlo Algorithm Approach. The major reason behind this project was to come up with conducive way which saves resource limited farmers of sheep in being charged exorbitant prices in trying to use the more advanced technology such as (DNA) in characterizing indigenous sheep. Sun, 01 Jan 2017 00:00:00 GMT https://cris.library.msu.ac.zw//handle/11408/3714 2017-01-01T00:00:00Z Chipindu, Lovemore