University of Zambia Online Research Repository
Not a member yet
    8755 research outputs found

    Multi-level student peer assisted mentoring (mspam) of computer science female students at undergraduate.

    Get PDF
    Thesis of Master of Science in ComputerThe percentage of female students majoring in computer science is lower than that of male students. Computer science has one of the lowest shares of women degree recipients among the broader field of Science and Engineering, with reports of only nineteen percent of the computer science degrees in 2016 been awarded to women. As such, fewer women than men enter the computing profession, leading to a gender disparity in the computing workforce. One way to improve the current state is to mentor female students in the field of computer science at both undergraduate and in industry. This study explored the current state of mentorship programs in institutions of higher learning in Zambia with the aim to develope a structured mentoring program for enhancing the performance of undergraduate computer science female students. The study used interviews and observations to gather qualitative data in order to review mentorship program that the universities are running for the female computer science students at undergraduate. The study found that the informal type of mentorship was used by 35 percent of individual lecturers without being aware of its application and structure. After reviewing the mentorship programs in the universities the results show that mentorship programs were 100 percent non-existent in the Zambian universities. While mentoring was something that would improve the performance, it still however, presented these two challenges, lack of clear definition for mentorship and structuring the program to be used to improve the number of female computer science students. This study has led to Multi-level Student Peer Assisted Mentoring (MSPAM) framework for a 4 year computer science undergraduate program. MSPAM optimises peer mentoring by drawing mentors from other students in different years of learning and organising these mentors into different levels of mentorship. The lower female graduation rates in the field of computer science contribute to continued disparities in which women only account for nineteen percent of the computer science workforce. Offering Multi-level Student Peer Assisted Mentoring would help close the gap

    An intergrated assessment tool for the development of green low-cost houses in Zambia.

    Get PDF
    ThesisGreen building tools (GBTs), used for planning and assessing the impact of structures on the environment, are developed in consonance with user needs. To attain green low-cost houses, consideration of the three arms of sustainability, i.e., environmental, economic, and social factors should be included. The aim of the study was to investigate the knowledge and utilization of GBTs among the professions in the building industry and to explore the experience of the respondents in applying green strategies in their housing projects. The study identified factors associated with the development of green low-cost houses and suitable criteria for assessing them. The results of the study were used to develop a green building planning and assessment tool for low-cost houses. An integration of criteria in selected tools, parameters in the Sustainability Development Goal 11 and the Sustainability Housing Guidelines were incorporated in the tool. The pragmatic paradigm approach was adopted and epistemologically the research was both subjective and objective in nature. This permitted the use of a mixed method design. An exploratory sequential mixed method approach was adopted. The data was collected using a questionnaire survey and interviews. The results of the study revealed that utilization of GBTs was low, and the knowledge of green buildings varied across the study population with architects having the highest knowledge. The factors that were associated with the use of GBTs were years of experience, the cost of assessment, environmental benefits, lack of technical knowledge, the availability of laws and regulations and tax reduction on green buildings and materials. Environmental criteria were preferred over social or economic factors and these included energy use, materials and health and well-being. The study findings showed that some of the criteria in GBTs were inappropriate for the Zambian building industry, and the expense of certifying structures made the tools unaffordable. The developed tool was made up of 8 components: environmental; social; economic; educational; management; regional priority; community participation and innovation. It had 10 assessment criteria and 16 different planning, design and material solutions integrated into a computer-generated program. The tool was evaluated by experts and recommended for use in updating the current National Housing Policy and architects’ practice manuals. It can also be used for dissemination of green building knowledge to developers, communities, and local planning authorities. Keywords: Rating Tools, Green Tools, Criteria, Sustainability

    School of Agriculture Science:Past Examination Papers,2020/2021

    No full text

    Factors associated with the increase in bottle feeding among children 0-5 months in Zambia.

    Get PDF
    Thesis of Master of Public Health.The study aimed at determining factors associated with the increase in bottle feeding among children 0-5 months in Zambia. The Data used in the study was drawn from a cross sectional study design collected through Demographic Health Survey and was analyzed using Stata version 15.0. To investigate the association between the dependent variable and the independent variables, a simple binary logistic regression analysis was carried out. Then multivariate logistic regression analysis was used based on a significance level of p-value < 0.05. Thus the adjusted odds ratios were obtained by entering at once all the independent variables that were found to be significant in the bivariate analyses. The study was guided by the following research questions. a) What is the magnitude of bottle feeding? b) What are the socio-economic factors associated with bottle feeding? The study revealed that, 3.3% of women bottle-fed their children while 96.7% did not. The study found that women with higher education had the highest percentage of bottle feeding at 21.48% followed by the richest women at 13.92% and 11.45% for women that delivered by caesarean section. It was found that richest women had higher odds of bottle feeding than the poorest women with (AOR = 7.92; 95% CI: 3.967, 15.806). It was revealed that women who delivered by C-section were1.51times more likely to bottle feed than those who did not deliver by C-section. On the other hand, Women from rural areas were 0.66 times less likely to be bottle feed compared to those in urban areas. Therefore, the study recommends: i. The establishment of friendly breastfeeding workplace policies for working mothers. ii. The enactment of laws to allow for six months paid maternity leave to enable employed mothers to exclusively breastfeed. iii. For strict adherence to ARV treatment for breast feeding parents living with HIV. iv. The need for future research on the effectiveness of interventions aimed at improving exclusive breast feeding

    Investigating factors associated with continued urban poverty in Zambia: evidence from Kalingalinga township, Lusaka.

    Get PDF
    Thesis of Master of Business Administration - FinanceThis Study is an inquiry into the factors associated with continued urban poverty in Zambia a case of Kalingalinga Township of Lusaka. Poverty in the context of this study is defined as a circumstance of serious deprivation where a person lacks one or more basic need as opposed to a condition of inequality. A mixed method design was used in carrying out the research with a sample of 40 respondents purposively selected. A self-developed semi-structured questionnaire was used in primary data collection. In analyzing the data, SPSS was used for quantitative data and thematic analysis for qualitative data. The findings indicate that poverty is significantly correlated (all sig. ≤ 0.01) with social factors such as natural causes (r = 0.365), bad choices (r = 0.331) and enablement (r = 0.236). The research further concludes that the poverty in Kalingalinga, was largely associated with psychological and economic factors. The findings also review that the people of Kalingalnga have developed poverty coping mechanisms such getting involved in village banking. For the government interventions, it is revealed that Social Cash Transfer and CDF are the main poverty interventions. The research ends with a recommendation of a policy framework for managing urban house hold poverty in Zambia

    a case study of Kwame Nkrumah University, Zambia

    Get PDF
    ArticleOne of the core principles of Kwame Nkrumah University (KNU) is training for transformation with a primary goal of decolonizing students mind set to offer transformative service to the nation and the society at large. This means decolonizing the thoughts, preferences and values that derive from a colonial way of thinking. A number of scholars have focused on the value of transformation with little regard on how far the curriculum can help decolonizing students’ mindset to offer transformative service to the nation and the society at large. This is the knowledge gap this study addressed. The objective of the study was to suggest strategies that can be employed to decolonize students mind set to offer transformative service to the nation and the society at Large. The study employed Case Study design. The study was guided by the transformative theory of leadership. The sample under the study consisted of 8 lecturers, 20 students, 3 Heads of departments, 1 Assistant Dean Postgraduate studies. Others included 1 Assistant Director postgraduate studies and a Registrar for ODL. Convenience sampling was used to select Kwame Nkrumah University. Purposive sampling technique was used to select lecturers. The Post-graduate students were selected using the quota sampling. The researchers used questionnaires with open-ended questions and semi- structured interview guide. Thematic analysis was used to analyse open-ended questions and semi-structured interviews. The results indicated that an educator should be a role model to what is expected of a student. The curriculum needs for a paradigm shift to vocationalise our curriculum. The knowledge acquired by students should help the nation and the society at large to be critical thinkers, innovative and productive hence decolonization. Keywords: decolonization, mind-set, transformative service, nation, societ

    Petrogenesis and potential for magmatic mineralisation of the mafic dykes of the northeastern part of the Irumide fold belt of Northeastern Zambia.

    Get PDF
    Thesis of of Doctor of Philosophy in GeologyThe area of northeastern Zambia, between Chama and Lundazi towns, is underlain by rocks belonging to the Irumide Fold Belt, a Mesoproterozoic orogenic belt whose geology is dominated by biotite gneisses, which often show incipient migmatisation, that belong to the Kampemba Group. Less common unmigmatised rocks types are schists, quartzo-feldspathic gneisses, calc-silicates, quartzites and metamorphosed basic and aplitic sills. Porphyroblastic leucogneisses are represented as narrow strips and these have general NE-SW trending foliations, the general Irumide structural fabric, and dipping to the NW. A complex of a mafic dyke swarm that occupies a NE trending fracture zone, occurs mainly between Chama and Lundazi. Individual dykes vary in size from about 100s of metres long by a few 10s of metres wide, to about 4 km long and about 100 metres wide. By nature of their emplacement, the dykes could contain potential for economic magmatic mineralisation. This study, therefore, looked at the petrogenesis of the mafic dyke swarm, through the application of whole rock geochemistry, geochronology and isotope geochemistry. The study also made an assessment of the potential for Cu-Ni-PGE sulphide mineralisation of the dyke swarm by applying chemical data and empirical evaluation methods. Fifteen fresh samples were sent to Actlabs in Canada for whole rock geochemical analysis. Three of the fifteen samples were analysed for Rb-Sr and Sm-Nd isotope geochemistry; and the same three samples were used in the K-Ar age determination of the rocks. The results received on age determination gave the ages, 1522 ± 90 Ma, 1067 ± 55 Ma and 608 ± 35 Ma. Sm and Nd concentrations range from 1.99 – 3.112 ppm and 6.666 – 9.88 ppm respectively. ENd were calculated by the laboratory at -5.23, -8.23 and -9.73. 87Sr/86Sr returned values of 0.710886, 0.714503 and 0.719612. From the study of the thin sections and photomicrographs, it has been concluded that the dykes studied are gabbroic in composition and exhibit differences in composition, textures, alteration and microstructures. They are composed mainly of plagioclase, clinopyroxene, hornblende with subordinate olivine, quartz, sericite, epidote and opaque phase minerals. The dyke rocks occur mainly in coarse and medium grained varieties as only two of the twelve samples examined, are fine grained. Several diagrams which have been applied in the evaluation of crystallisation fractionation of both types of rocks have revealed that fractional crystallisation affected the magma of both types of rocks studied, tholeiitics and calc-alkalics. This is further supported by the wide ranges of Mg#, MgO, Ni and Cr in both rock types, which indicate significant crystallisation fractionation either in magma chambers or during ascent. The low Mg# (29.7-58.0), Ni (80-720 ppm) and Cr (150-1400 ppm) are suggestive of evolved magmas that underwent fractional crystallisation. Since no two samples came from the same dyke, it can also be concluded that all the mafic dykes in the Study area experienced various levels of fractional crystallisation. Both types of dykes show enrichment of the source magma by addition. The enrichment must principally have come from the terrigenous felsic sediments rich in HFSE and LILE which were subducted at the mantle wedge. From the analysis of all the available information that would be critical to sulphide mineralisation and possible occurrence of deposits of Cu-Ni in the mafic dykes of the study area, it is concluded that the dykes exhibit no parameters that would have allowed such mineralisation to take place. Whilst the study has revealed, from age dating perfomed on three samples, that the mafic dykes of the northeastern part of the Irumide Fold Belt of northeastern Zambia were emplaced in at least three pulses, it is still recommended that high precision and more expensive age dating methods such as U-Pb Zr are carried out

    Assessment of the implementation process of the electronic health commodity management information system in Lusaka district health facilities.

    Get PDF
    Thesis of Masters in Business AdministrationOver the past decade, the Ministry of Health (MOH) in Zambia has experienced health commodity inventory management challenges that resulted in stock outs, expiry and pilferage of life-saving medicines. A weak Logistics Management Information System (LMIS) was cited as a major factor causing the inability of Supply Chain (SC) users to make sound data-based SC decisions. An evaluation of the paper-based LMIS attributed the lack of accurate, goodquality data captured and reported at Service Delivery Points (SDPs) to the laborious nature of the paper-based manual processes that required human resources, time and stationery. To address this challenge, the recommendation was to automate the system by implementing the electronic Logistics Management Information System (eLMIS). The eLMIS was envisaged to ease the data capture and reporting processes at SDP leading to an increase in data visibility, access and data-based SC decision making. The SC managers would at the press of a button have readily available supply status updates to aid aversion of stock imbalance and be equipped to plan re-supply. Currently, the facility level eLMIS is rolled out to 51% (1,643/3200) of public health SDPs while the central edition is rolled-out countrywide. While the reporting rates are in the ninetieth percentile, SDPs continue to experience stock imbalances, expires and instances of pilferage (eLMIS reports, 2023). A key question: Working on the assumption that the design of the LMIS is ideal, how effective was the rollout process of the eLMIS executed? Otto et. al., (2015) in a discussion paper focused on Africa, explains the five pillars of health systems noted to have a direct impact on the success of ICT innovation implementation in Africa. These foundational elements are infrastructure, governance, healthcare financing, supply chain management, and human resources for health. Building on the five pillars, to successfully establish and scale up an automated LMIS, the critical prerequisites are data and interoperability standards, local capacity, policy and regulatory environments, an appropriate business model, alignment of partnerships and priorities, and monitoring and evaluation. While some prerequisites are crosscutting, there is a need for further refinement of other prerequisites depending on the specific country context. This study set out to establish if the implementation process for the eLMIS is yielding the desired system and end-user outcomes. The study evaluated the prerequisites outlined by Otto et. al., (2015) by engaging key stakeholders at all levels of the SC in managing health commodities and supporting implementation. While the findings noted important successful attributes, several significant issues include threats to sustained operations resulting from leadership and policy implementation gaps, limited dissemination of critical information among stakeholders, lack of clarity of roles among eLMIS stakeholders, funding resource challenges, and donor dependence

    Socio-ecological drivers and dynamics of limnothrissa miodon on lake Tanganyika: a case of Mpulungu district.

    Get PDF
    Thesis of Master of Science in Environmental and Natural Resources Management.The dissertation examined drivers that contribute to the decline of Limnothrissa miodon on Lake Tanganyika. Explicitly, the research study pursued; (i) factors that contribute to the reduction of Limnothrissa miodon, (ii) effects of not sustaining Limnothrissa miodon on fishers and marketeers and (iii) strategies that could be used to sustain Limnothrissa miodon. The case study research design targeted fishers, marketeers, and key discussants. Qualitative methods were used to capture the attitude, views, ideas and opinions of respondents on the reduction of Limnothrissa miodon. Snowball sampling, and expert purposive sampling techniques were used. One hundred and twenty-four (124) respondents were sampled through theoretical saturation in Kasasa, Kabatwe, Musende fishing villages and Ngwenye market. Questionnaires, Focus Group Discussions (FGDs), and interview schedules were used as data collection instruments. Thematic analysis was adopted to make codes, and give meanings. Ninety-two (92) respondents representing seventy four per cent (74%) attributed indiscriminate fishing to the reduction of Limnothrissa miodon. Twenty-seven (27) respondents representing twenty-two per cent (22%) attributed the increase in the population of fishermen, and marketers to the reduction of Limnothrissa miodon. On the other hand, five (5) respondents representing four per cent (4%) attributed climate change to the reduction of Limnothrissa miodon. This study, subsequently, recommends the introduction of fishing quotas, bag limits, fisher licensing, closed seasons, size limits, aquaculture practices, creation of freshwater reserves, and freshwater-protected areas. The study suggests further research on the prevalence of wrong fishing gear, and illegal fishing methods

    Detection of android adware using transfer learning with computer vision.

    Get PDF
    ThesisAndroid malware is a dominant threat category for consumers with a vast majority of this malware being Adware. Adware is a type of malware that displays unsolicited pop-up advertisements that are used to generate revenue and can also lead to installation of trojan horses, spyware, and other malicious software. Over the recent years there has been an increase in the number and complexity evasive malware, especially Adware. Despite this, there has not been a lot of focus put on the Adware malware family. Malware authors are using obfuscation techniques to make their malware difficult to detect using conventional static and dynamic methods of analysis, this has led to the rapid adoption of artificial intelligence and machine learning approaches to overcome these challenges. Malware binaries can be visualized as images, with the observation that for many malware families, the images belonging to the same family appear very similar in layout and texture. These images can then be used as input to train deep neural networks to detection malware. The goal of this research is to detect android Adware using visual representation of an android application package (APK) Dalvik Executable (DEX) file and providing it as an input image to a pre-trained neural network can successfully detect android Adware apps. Using the CICMalDroid2020 dataset obtained from the University of New Brunswick Canadian Institute of Cybersecurity as training, validation and test datasets for my experiments, the DEX files in each APK file are first extracted from the APK files and then converted to grayscale images. The obtained images are then applied as input to deep neural networks that use selected pre-trained models namely VGG16, ResNet50, InceptionV3, EfficientNet v2 and MobileNet v2 to detect android Adware. Using the confusion matrix to calculate each model’s accuracy, recall, precision, and f-1 score, the performance results of each trained model are compared amongst all the models used in this research. The highest performance measurement of 92 % detection rate on the f1- score performance metric was achieved by the MobileNet v2 model. The results obtained in my research reveal that computer vision and transfer learning perform adequately in the detection of android Adware apps from benign apps. Keywords: Android Adware, Computer Vision, Deep learning, pre-trained model, Convolutional Neural Network (CNN), Transfer learnin

    4,646

    full texts

    8,755

    metadata records
    Updated in last 30 days.
    University of Zambia Online Research Repository
    Access Repository Dashboard
    Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇