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 Comprehensive Review of System Log Significance and Preparation Techniques for Experimental Studies
(International Journal of Scientific Research in Computer Science and Engineering, 2025-12) Muchori, Juliet Gathoni; Mwangi, Peter Maina
System logs are essential assets in today’s experimental research supporting trends like AI driven analytics, automation and reproducible science. By providing preparation techniques, the review has provided a unified framework that is able to enhance data quality, improve experimentation and increase system complexity. This later assists in developing more accurate, more accurate, scalable and futuristic models, especially in big data, machine learning and intelligent systems. This review aims to systematically find out the importances of system logs in research and to find out how logs are prepared before they are used for research. Research papers dated 2021 to 2025 were used in the review. Barbara Kitchenham’s guidelines were used to review the journal papers and to write the entire paper. Several databases were used to get the literature such as IEEE, Elsevier, MDPI, Springer, SpringerOpen, Wiley, Frontiers, PLOS, PeerJ, Taylor & Francis, arXiv (preprint). The results indicated that importance of logs included doing objective event capture by use of time stamps, it’s aground of truth for security or anomaly experiments, it’s an input in ML experiments, acts as a provenience reproducibility artifact, it’s an experimental variable, its used in behavioural and interaction analysis, it used in performance monitoring and system optimization. It was also noted that logs preparation is done in stages that is, define scope and objectives, collect and centralize, normalize timestamps, parse logs, clean and canonicalize, feature extraction, sampling or instance selection, privacy or anonymization, validation and benchmarking, document and archive. For future works, researchers can create models advancing automated log analysis, standardization of log format and metadata, reproducible research frameworks
A Systematic Literature Review on Application of Text Embedding Techniques in NLP
(International Journal of Scientific Research in Computer Science and Engineering, 2025-12) Muchori, Juliet Gathoni; Maina, Mwangi Peter
In many natural language processing (NLP) tasks such as sentiment analysis, text classification, information retrieval, and topic modeling text embeddings or representation techniques are foundational. These methods convert raw text into numerical form so that machine learning or deep learning algorithms can work with them. Among the most widely used are TF-IDF a sparse, frequency-based representation, static embeddings like GloVe which capture word co-occurrence statistics globally but are non‐ contextual, and contextual transformer embeddings such as BERT, RoBERTa, and DistilBERT which encode words in context, enabling modeling of polysemy, syntax, and semantics in more dynamic ways. This review aims to systematically examine and compare existing models and studies from 2020-2025 that use these embedding/representation techniques individually or in hybrid form in classification-type tasks. Specifically, its objectives are: To review classification models using TF-IDF and assess their strengths and; To examine classification models that combine or use TF-IDF + GloVe /static embeddings, detailing how they improve semantic modeling but potentially suffer from interpretability or domain mismatch issues; To assess work with contextual embeddings (BERT, RoBERTa, DistilBERT), including their performance gains, computational/resource costs, and potential; Barbara Kitchenham guidelines will be used and journal papers to be review will be from scholar journal such as IEEE, Elsevier, Springer, ACM Digital library, Citeseer Library, arXiv, Wiley. It was noted that TFIDF suffers from Sematic meaning of words which can be solved by Glove, BERT, Word2Vec or GloVe. For future works there is one promising direction which is to combine TF-IDF with semantic embedding methods, such as word embeddings or contextual embeddings like BERT, to capture both lexical and contextual features of text.
Optimized Solutions for Robust and Efficient Two-Factor Authentication in Networking Environments
(Babylonian Journal of Networking, 2025-12) Hussein Alkattan; Raad S. Alhumaima; Amr Badr; Peter Mwangi
With the rise of escalating cyber threats to the present-day networking environment, the traditional two-factor authentication (2FA) mechanisms remain ineffective in mitigating the sophisticated attack vectors like phishing, SIM-swapping and social engineering. These loops holes, specifically in SMS- and email-based 2FA, are opening users and network infrastructure up to substantial danger. This paper presents optimized and robust enhancements to 2FA technology, and points out cryptographic technologies like ECC, the addition of X.509 digital certificates, and biometric and behavioral authentication solutions. Then, the performance of these distributed trust models is compared, in terms of security efficiency, usability, and deploy ability, with a complete comparative study of these models in dynamic networking. The paper also includes real-world examples of implementing such multi-layered 2FA schemes being tensioned between a strong security protection and what is deemed to be user acceptable. The results showcase best practices and challenges in building secure and efficient as well as future-proof authentication systems that match the requirements of complex network environments.
CLUSTER HEAD SELECTION ALGORITHMS FOR ENHANCED ENERGY EFFICIENCY IN WIRELESS SENSOR NETWORKS: A SYSTEMATIC LITERATURE REVIEW
(International Journal of Computer Science and Engineering Survey (IJCSES), 2022-06) Mwangi, Peter Maina; Ndia, John Gichuki; Muketha, Geoffrey Muchiri
The extension of the sensor node's life span is an essential requirement in a Wireless Sensor Network. Cluster head selection algorithms undertake the task of cluster head election and rotation among nodes, and this has significant effects on the network's energy consumption. The objective of this paper is to analyze existing cluster head selection algorithms and the parameters they implement to enhance energy efficiency. To achieve this objective, systematic literature review methodology was used. Relevant papers were extracted from major academic databases Elsevier, Springer, Wiley, IEEE, ACM Digital Library, Citeseer Library, and preprints posted on arXiv. The results show that there are many existing Cluster Head Selection Algorithms and they are categorized as deterministic, adaptive and hybrid. These algorithms use different parameters to elect Cluster Heads. In future the researchers should derive more parameters that can be used to elect cluster heads to improve on energy consumption.
An Extended Low-Energy Adaptive Clustering Hierarchy Routing Protocol for Efficient Energy Consumption in Wireless Sensor Networks
(International Journal of Research in Advent Technology, 2024-06) Mwangi, Peter Maina; Ndia, John Gichuki; Muketha, Geoffrey Muchiri
In wireless sensor networks (WSNs), a primary objective is to extend the lifespan of sensor nodes. Cluster head selection algorithms play a crucial role in electing and rotating cluster heads among nodes, significantly impacting the network's energy utilization. Over the years, various energy-efficient routing protocols have been developed to reduce energy consumption and thereby prolong the network's lifespan. Current energy-efficient routing protocols, such as HEED, TEEN, APTEEN, SHPER, and LEACH, have not fully addressed the challenge of energy consumption in WSNs. LEACH, which stands for Low Energy Adaptive Clustering Hierarchy, is a well-known clustering protocol designed for energy-efficient data gathering in WSNs. However, the processes of selecting cluster heads and the effectiveness of data aggregation in the basic form of LEACH can be complex. This study aims to develop an extended version of the Low Energy Adaptive Clustering Hierarchy routing protocol that employs an extended K-Means Cluster Head Selection Algorithm to choose cluster heads more effectively. The developed protocol is intended to enhance the longevity of WSNs. A quantitative approach has been utilized to measure performance by simulating various routing protocols. To demonstrate the advantages of the proposed protocol, we compared it against previous protocols using several metrics, including residual node energy, packet delivery ratio, throughput, network longevity, average energy consumption, and the number of live and dead nodes. The results indicate that the proposed protocol outperforms existing protocols, such as LEACH and SEP.