AFRICAN JOURNAL OF SCIENCE, TECHNOLOGY AND ENGINEERING (AJSTE)

ENHANCING THE CYBER RESILIENCE OF INFORMATION SYSTEMS THROUGH MACHINE LEARNING.

2026-10-26T12:28:32+01:00

Organizations today operate in an environment of relentless cyber threats that jeopardize continuity of critical services. Conventional security approaches, primarily focused on preventing intrusions, have
proven insufficient. This shortfall has shifted attention to cyber resilience, defined as organizations' capacity to maintain operations and recover effectively after cyber disruptions. Among the technologies
driving resilience, Machine Learning (ML) stands out for its ability to analyze vast datasets, identify anomalies, and respond effectively to evolving threats. Its ability to process vast and complex datasets
enables detection of subtle, hidden patterns that conventional tools or human analysts might overlook.To examine ML’s role in strengthening cyber resilience, this study undertook a systematic review of
recent literature (2020–2026). The review encompasses 87 peer‑reviewed studies from reputable sources, with the aim of mapping ML applications across different stages of resilience and identifying recurring
benefits and challenges. The goal is to provide a clear understanding of how ML contributes to resilience, evaluate its performance in practice, and identify the key challenges that must be overcome to ensure its
safe and robust adoption. Results show that ML significantly enhances resilience by enabling proactive threat detection and adaptive response mechanisms. However, challenges such as data quality, model
robustness, and the need for human oversight limit its effectiveness. The evidence suggests that ML is not a standalone solution but rather an enabling technology whose success depends on thoughtful
design, integration with human expertise, and mitigation of inherent limitations. This paper contributes a structured framework for understanding ML’s role in cyber resilience and highlights critical issues
that warrant further investigation, offering a foundation for both practitioners and researchers to advance resilient cybersecurity strategies.

ARTIFICIAL INTELLIGENCE FOR SUSTAINABLE AGRICULTURE: ENHANCING CROP PRODUCTIVITY THROUGH MACHINE LEARNING IN KENYA

2026-10-26T13:09:02+01:00

Agriculture is the backbone of food security and economic development in many developing countries.However, smallholder farmers continue to face challenges such as climate change, low agricultural
productivity, and unpredictable weather patterns. Artificial Intelligence (AI) has emerged as a promising technology for improving farming productivity and efficiency. Despite its potential, adoption
of machine-to-machine (M2M) and AI technologies remain limited in sub-Saharan Africa due to inadequate infrastructure and technological constraints. This study examined the predictive value of
AI-driven modelling in yield forecasting and optimal potential planting schedules and whether or not machine learning (ML) algorithms can be used to minimize farming risks and enhance agricultural
output. The study analyzed weather patterns, soil characteristics, and historical crop yield data collected from 500 farms in Kirinyaga County, Kenya. A Random Forest regression model was developed to
predict optimal planting and harvesting periods. Model performance was assessed using the Mean Absolute Error (MAE) and the coefficient of determination (R²). Surveys and expert interviews were
conducted to identify barriers to AI adoption among smallholder farmers. The AI model demonstrated strong predictive performance, achieving an R² value of 0.91. Farmers who implemented the AIgenerated recommendations recorded an average crop yield increase of 27%. However, limited digital literacy, inadequate technological infrastructure, and insufficient support services were identified as major barriers to widespread adoption. These results demonstrate that AI-based decision support systems can significantly enhance agricultural productivity through improved resource utilization, and
climate resilience. Integrating AI with Internet of Things (IoT) technologies offers substantial potential for advancing precision agriculture. Future research should focus on developing integrated AI–IoT
farming models and formulating supportive government policies to accelerate AI adoption to optimize agricultural practices in developing countries.

SEMANTIC WEB AND LINKED DATA TECHNOLOGIES FOR NEXT-GENERATION LIBRARY KNOWLEDGE MANAGEMENT

2026-10-26T13:18:33+01:00

Academic and Research libraries are growingly through exponential rise of digital information, heterogeneous sources of data, and user prospects for seamless, intelligent knowledge access. Traditional
library knowledge management structures, usually siloed and record-centric, struggle with semantic discovery, interoperability, and amalgamation with the larger web. In response, Semantic Web and
Linked Data technologies have evolved as major innovations, allowing interconnected metadata, machine-readable, supporting dynamic organization of knowledge and advanced discovery services. To
understand this evolution, this study sought to understand adoption and efficacy of Semantic Web and Linked Data technologies in improving library knowledge management worldwide, regionally and in
Kenya. It examines the core technologies, current applications, limitations of traditional systems, and the semantic framework capacity to enhance interoperability, resource discovery, and service delivery.
Data was acquired using systematic literature review, analyzing 35 institutional reports, peer-reviewed journal articles, policy documents, and technical standards. Information was acquired from online
institutional repositories, academic databases, and platforms including Europeana, Google Scholar, and AfriLIS. Data was thematically synthesized across world, Africa, and Kenyan settings, highlighting
benefits, trends, challenges, and research gaps. Findings from the world show that libraries as evident in BIBFRAME, Europeana, and Wikidata integration shows enhanced interoperability, discoverability,
and semantic data reuse. Regionally, adoption is mainly pilot-driven, limited by technical skills, infrastructure, and funding. In Kenya, apart from Moi University with the pilot program, activities
remain largely training-oriented or conceptual, with limited practical implementation of Linked Data and RDF frameworks. The research highlights important gaps in operational implementation and
empirical evidence in Kenyan libraries. Semantic Web and Linked Data technologies have the capacity to improve library knowledge management through enabling machine-readable, interoperable
knowledge networks and advanced discovery services. Future studies should focus on capacity-building, pilot implementations, user-centered assessments, and cross-institutional integration to translate
theoretical structures into practical library solutions.

TEXT MINING AND TOPIC MODELLING FOR KNOWLEDGE DISCOVERY IN DIGITAL COLLECTIONS

2026-10-26T13:26:31+01:00

Kenyan research and academic libraries have been experiencing challenges managing the growing collections of digital scholarly publications making it hectic for librarians to systematically organize
resources and users to access relevant content efficiently. Text mining has evolved as a strong method for obtaining insights from large textual sets of data, allowing for thematic analysis, automated
knowledge discovery, and enhanced organization of content. Nevertheless, its usage in the Kenyan academic and research library context remains limited. This study explored text mining and topic
modeling application on scholarly publications using open access digital libraries. The study sought to identify research patterns, establish latent themes, and express how these techniques could enhance
accuracy in knowledge access, management of time, and thematic organization in the context of Kenyan academic and research libraries. The study involved collecting and preprocessing a corpus of library
science and digital library articles. This included tokenization, stop word removal, and standardization. Using word cloud visualization, prominent terms were identified, while TF-IDF analysis displayed
distinctive keywords. Dominant topics along with related document groupings were done using Latent Dirichlet Allocation (LDA), permitting insight into thematic constructs across the corpus. Analysis
showed existence of three major topics that include digital library systems, open access and repository management, and scholarly publishing practices. We identified frequent terms like “library,” “digital,”
“metadata,” and “repositories”, stressing on the use of text mining technology-driven library systems. The techniques expressed that text mining could systematically support automated classification,
display latent trends, and facilitate resource discovery in libraries. This study showcases the capacity of text mining to improve content organization and knowledge discovery globally, hence should be
practiced in Kenyan academic libraries. Future studies need to explore integration of AI driven text mining with library systems to further improve thematic analysis and user accessibility.