DSpace Repository

MAMA NGINA UNIVERSITY COLLEGE INSTITUTIONAL

REPOSITORY

MNUC repository is a digital service that collects, preserves, and distributes digital material. Repositories are important tools for preserving an organization’s legacy; they facilitate digital preservation and scholarly communication.

Communities in MNUC Repository

Select a community to browse its collections.

Now showing 1 - 5 of 5

Recent Submissions

A novel Elm optimization and its application in IoT intrusion detection
(Springer, Singapore, 2024-09-22) Maseno, Elijah M.; Wang, Zenghui
Globally, the Internet of Things (IoT) has grown exponentially in the last few years. The integration of numerous IoT devices enhances connectivity but also exposes them to various cyber-attacks. With the increase of complex and intelligent cyber-attacks, the traditional methods of cyber-attack detection, such as anomaly and misuse intrusion detection systems (IDS), have proven ineffective in combating emerging cyber-attacks. Scholars have suggested that the existing IDS’s performance can be enhanced by integrating machine learning methods. This study explored the integration of ELM (Extreme Learning Machine) with NSGA-II (Non-Dominated Sorting Genetic Algorithm II). Our empirical analysis reveals that ELM outperforms other algorithms regarding learning rate, usability, and effectiveness of generalization. NSGA-II provides an equilibrium between exploration and exploitation, successfully managing the trade-offs present in multi-objective optimization problems to identify a range of excellent solutions along the Pareto front. Most of the existing works on ELM optimization focuses on improving one input parameter. This study explored the integration of ELM with NSGA-II for both input weights and hidden neurons. The model was trained and tested using two datasets namely IoT_ToN network and UNSWNB15 datasets. This research used TON_IoT Windows dataset to further test the capacity of the model to identify novel attacks. Using the UNSWNB15 and IoT_ToN network datasets, respectively, the model’s accuracy was 0.68 and 0.65. In addition, the model recorded an accuracy 0.768 using TON_IoT Windows dataset, demonstrating the potential of the model in detection of new forms of attacks. This study acknowledges the ongoing evolution of challenges in IoT security and the need for continuous innovation to stay ahead of emerging threats in the future. The contributions of this study can be explored further to develop reliable intrusion detection solutions for the dynamic IoT landscape.
Role of community health workers monetary incentives on retention and health service delivery in Kibwezi district, Makueni county, Kenya
(2014) Mbugua, Ruth G.
The global policy of providing primary level care was initiated with the declaration of Alma-Ata in 1978. Kenya is a signatory to the Alma-Ata declaration. Implementation of the Community Health Services is a top priority for the Ministry of Public Health and Sanitation in Kenya. The second National Health Sector Strategic Plan (NHSSP II) defined a new approach to the delivery of Health Care Services to Kenyans, the Kenya Essential Package of Health (KEPH).CHWs are the key agents in the implementation of the community strategy. In Kibwezi District CHWs trained by MOPH&S do not receive monetary incentives while their counter parts trained by other partners (AMREF, USAID-APHIA II and USAID APHIA plus) receive monetary incentives. The study was done to find out the effect of monetary incentives on retention and performance of Community Health Workers in Kibwezi District in Kenya. A Cross-Sectional Comparative study design was used for the study. Qualitative data was collected through Key Informant Interviews and Focus Group Discussions were also conducted, one comprising of Community Health Committee members. Quantitative data was collected by the use of a structured questionnaire. Multi stage, purposive and simple random sampling were used to select 4 Community Units receiving incentives and 4 Community Units not receiving monetary incentives for comparison purposes. A total of 282 CHWs were interviewed 140 from Community Units receiving monetary incentives and 142 from CUs not receiving monetary incentives in Kibwezi District. Chi-square was used to establish the relationship between the research variables. Association between the variables was analyzed using chi-square tests and cross tabulations. Data was presented in form of figures, tables and narration. Age, [OR 3.6327 P= 0.022], marital status [OR 3.306 P= 0.018], education level.[OR 2.901786 P= 0.002], and occupation [OR 2.901786 P= 0.002]were significantly associated with performance of CHWs. Subsequent training[OR =2.7469, P value= 0.008], supervision [OR =5.95522, P= 0.0001], training partner [OR 3.97, P= 0.023]were significantly associated with performance. CHWS receiving monetary incentives were better performers. There was a significant difference in the number of women referred for antenatal care (P =0.022), number of women with newborns who had been counseled on exclusive breastfeeding (P =0.043) and the participation of CHWs in community dialogue days. (P=0.005) between the two groups. CUs receiving monetary incentives had better key health indicators in CUs receiving monetary incentives. There was a significant difference in the proportion of children below 5 years who were fully immunised (P= <0.0001), proportion of women who had attended 4 ANC visits (P=0.028) and the proportion of pregnant women delivering with a SBA. (P=0.003).CUs not receiving monetary incentives had higher attrition rates of CHWs (13%) than CUs receiving monetary incentives(4%).(P=0.013).There is a need for government and partners to explore sustainable perfomance based financial incentives which will ensure all the CHWs receive monetary incentives. Findings from this study will be used by the policy makers as a guide to decision making on improvement of performance and retention of CHWs and which will in turn improve health indicators of the communities at large.
Tackling the cancer literacy needs: review findings from Africa and the way forward
(Springer, Singapore, 2025-02) Harsch, Stefanie; Kugbey, Nuworza; Mbugua, Ruth G.; Sørensen, Kristine
It is a verifiable fact that the incidence of cancer is on the rise on a global scale, including in Africa. A high level of cancer literacy is essential for patients, their families, and healthcare professionals to be able to undergo cancer screening, cope more effectively with a diagnosis, adhere to a treatment plan, and manage the disease more successfully. To date, there is a paucity of knowledge regarding cancer literacy on a global scale, and it has never been examined for the African continent as a whole. This chapter aims to illustrate the importance of cancer literacy, define it, and review the various survey instruments that have been employed in this field. Subsequently, the chapter presents an overview of cancer literacy in Africa, based on existing studies, and discuss the challenges and initiatives at all levels of society to enhance cancer literacy. In conclusion, the chapter presents a discussion of the way forward. The chapter illustrates the critical importance of cancer literacy and the necessity of collaborative initiatives to enhance cancer literacy at the political, organizational, community, and individual levels. This overview chapter and the concrete examples it presents may serve as a source of inspiration for this endeavor.
A systematic review on hybrid intrusion detection system
(Wiley, 2022-05) Maseno, Elijah M.; Wang, Zenghui; Xing, Hongyan
As computer networks keep growing at a high rate, achieving con.dentiality, integrity, and availability of the information system is essential. Intrusion detection systems (IDSs) have been widely used to monitor and secure networks. +e two major limitations facing existing intrusion detection systems are high rates of false-positive alerts and low detection rates on zero-day attacks. To overcome these problems, we need intrusion detection techniques that can learn and e8ectively detect intrusions. Hybrid methods based on machine learning techniques have been proposed by di8erent researchers. +ese methods take advantage of the single detection methods and leverage their weakness. +erefore, this paper reviews 111 related studies in the period between 2012 and 2022 focusing on hybrid detection systems. +e review points out the existing gaps in the development of hybrid intrusion detection systems and the need for further research in this area.
Effects of mobile health technologies on uptake of routine growth monitoring among caregivers of children aged 9 to 18 months in Kenya
(2021) Nyang'echi, Edna; Osero, Justus
This study aimed at finding out the effects of mobile health (mhealth) technologies on uptake of Routine Growth Monitoring (RGM) among caregivers of children aged above 9 months in Kenya. This was a quasi-experimental study. The experiment groups received Short Text Message (STM) and Voice Call (VC). The analysis demonstrates that in month 1, caregivers who received STM were 6.875 times more likely to take their children for RGM compared to control (OR = 6.875; 95 CI: 3.591-13.164); caregivers who received VC were 6.750 times more likely to take their children for RGM compared to those in control arm (OR = 6.750; 95 CI: 3.522-12.938). Policy makers and implementers in the health will find these study findings useful in deciding whether or not to adopt STM or VC in improving uptake of RGM for children above 9 months.