Salem Journal of Science, Information & Communication Technology

The most commonly used paradigms for deep learning optimization are based on localized summary point-estimates, e.g., validation accuracy, precision and recall, and F₁-score for convolutional neural networks (CNNs). These measures are good indicators of the "central" component of model predictive ability, but are systemically unresponsive to the dispersion of the process and high-frequency variability in its predictive power associated with stochastic initialization noise and the concurrent variability of the hardware. Therefore, models trained with mistaken high accuracy in training/test sets are very susceptible to operational degradation if deployed in a distributed edge setting. This paper tries to fill this gap, offering a comprehensive bibliometric and systematic literature review (SLR) to evaluate the structural integration of quality engineering (QE) principles into the machine learning lifecycle. A bibliometric corpus of 1,142 publications identified from the Scopus database (2015–2026) was clustered and classified using keyword co-occurrence and thematic mapping, indicating that computer science and quality engineering studies have traditionally existed in parallel communities with very little convergence. Moreover, a separate PRISMA-screened SLR corpus of 1,892 raw candidate records was subjected to stringent exclusion criteria. This reduces the sample to just 55 high-fidelity studies (2.9% retained). In essence, this demonstrates that transferring a process capability level to the active training loops using a statistical process capability (Cpk)-level is still extremely rare [19]. Finally, this paper builds on these quantitative empirical gaps to suggest a conceptual synthesis, which is implemented in a structural way for AI development. We translate algorithmic misclassifications into process defect rates, and set Individuals and Moving Range (I-MR) process control limits to offer a structured roadmap to re-optimizing deep learning as a continuous process that is statistically controlled, like other industrial manufacturing processes. [18].

There is a cardiac crisis in Nigeria - less than 100 cardiologists for a population of greater than 220 million and more than 200 000 out of every 300 000 people in the country die from cardiovascular disease. Although progress has been made in cardiovascular AI research in the past decade, there is currently no machine learning model that can diagnose heart disease from data of Nigerian patients. This paper outlines an ML-CDSS that is specifically targeted to the Nigerian tertiary care hospitals. The CDSS has six main structured clinical features and raw 12-lead electrocardiogram (ECG) waveforms and adopts a multi-modal stacked ensemble approach with XGBoost, feedforward neural network, as well as a CNN-LSTM deep learning sub-model. In addition to each prediction, TreeSHAP provides an explanation layer for the tree-based approach, which is easy to understand, and DeepSHAP yields an explanation layer for the neural network; LIME is used to pass on secondary explanations, and GradCAM to visualise the activations for the ECG waveform. The process is done through five phases, HEART (Harvest, Engineer, Architect, Refine, and Transform) including prospective data collection from three Nigerian federal tertiary care hospitals, pre-processing, modelling building, validation with a holdout set blinded data, and deployment. Architecture designed to run intermittent connections, commodity compute, energy constrained. Since the inception of this project, there have been implicit collaboration between governance and NAFDAC Software to develop a Medical Device guide and Nigeria Data Protection Regulation guidelines. This paper tackles seven known gaps within the cardiac AI literature in Africa head-on.

The escalating demand for sustainable energy has driven the exploration of waste-derived feedstocks and catalysts for biodiesel production. This study investigates the optimization of biodiesel yield from waste cooking oil (WCO) using a novel eggshell-anthill mixed oxides composite catalyst via the Taguchi L9 orthogonal array design. Four key process parameters methanol-to-oil molar ratio (8:1–12:1), catalyst weight (1.5–2.5 wt.%), reaction temperature (60– 100 °C), and reaction time (40–60 min) were evaluated. The catalyst was prepared by blending eggshell and anthill powders in a 1:1 ratio, followed by calcination at 850 °C for 4 hours. Characterization techniques (XRF, SEM, and XRD) confirmed high CaO content (72.85 wt.%), porous irregular morphology, and crystalline CaO phases suitable for transesterification. Experimental results showed biodiesel yields ranging from 80.76% to 90.87%, with the regression model exhibiting excellent fit (R² = 0.9878). ANOVA and contribution analysis revealed catalyst weight as the most influential factor (85.42% contribution), followed by reaction time and methanol-to-oil ratio. Optimal conditions were identified as methanol-to-oil ratio of 10:1, catalyst loading of 2.0 wt.%, and reaction time of 40 min, yielding a maximum of 91.23% FAME. The produced biodiesel met ASTM D6751 and EN 14214 standards in key physicochemical properties. This work demonstrates the potential of low- cost, dual-waste-derived heterogeneous catalysts for efficient and environmentally friendly biodiesel production.

Analysis of transfer ration of radionuclides concentration of Water to Fish (Bonga Shad) and Sediments to Fish (Bonga Shad) samples of the Coastal communities of Okrika, in Rivers State, which lies between latitude: 40.43'44" N and 40.45'57" N and ranged from longitude: 70.3'20" E to 70.6'42" E. Nigeria using gamma spectroscopy method with NaI (TI) detector. The transfer factor of radionuclides from for Sediments to Fish, ranged from 1.28 - 21.57 with a mean value of 8.35± 6.61, 0.19 to 5.74 with a mean value of 2.37 ± 2.60 and 1.49 -5.74 with a mean value of 2.82 ±1.29 for 40K, 238U and 232Th respectively, and the magnitude of the transfer factor is of the order 40K > 238U > 232Th. The Transfer factor of radionuclides from the Water to Fish ranged from 0.77 to 3.79 with a mean of 3.79 ± 2.05, 0.26 to 47.73 with a mean 6.58 ±15.52 and 0.94 – 3.43 with a mean value of 2.02 ± 0.90 for 40K, 238U and 232Th respectively, and the magnitude of the transfer factor is of the order 238U > 40K > 232Th. These results obtained are all higher than the stipulated reference point of 1.0 which indicates that bioaccumulation of radionuclides exist in the Fish (Bonga Shad) samples from the intake of radionuclides from water and sediments of the Coastal communities of Okrika. 

Dual contraceptive use is a critical approach to preventing unintended pregnancies and reducing the transmission of HIV and other sexually transmitted infections (STIs). Despite their significance, knowledge, and utilisation of dual contraceptive methods among women living with HIV in Oyo State, Nigeria. This study examined the extent of awareness regarding dual contraceptive utilisation among sexually active women with HIV in Oyo State. The study applied a descriptive research design and employed a structured questionnaire to gather data from respondents receiving treatment at specific health facilities delivering ART services in Oyo State. Participants were selected through systematic random sampling at three health facilities: Adeoyo Maternity Health Centre, Lad Medical Centre, and St. Anne's Hospital. Descriptive statistics with the aid of IBM SPSS Software application were employed to evaluate the respondents' comprehension of the advantages and functions of dual contraceptive methods. The study's results indicated that respondents' levels of knowledge were inconsistent. Although a significant number of participants were aware that dual contraceptive methods can prevent HIV transmission to sexual partners and reduce the risk of other sexually transmitted infections, their understanding of other aspects of dual protection was restricted. The suggestion that women living with HIV should combine condom use with other modern contraceptive methods was met with significant disapproval by a substantial number of respondents. The study concludes that sexually active women living with HIV in Oyo State are aware of some of the benefits of dual contraceptive use; however, comprehensive knowledge is still lacking.

Now a day’s there is an increasing recognition that freshwater is a valuable resource due to overexploitation and pollution. Wastewater discharge contains several harmful substances or chemicals, which may cause adverse environmental impacts such as changes in aquatic habitats, species composition, and a decrease in biodiversity (Li, et al., 2024). All of these impacts lead to a less valuable environment, a less prosperous economy, and ultimately, a diminished quality of life. Recent studies have highlighted the complex interactions between multiple pollutants in sewage and their synergistic negative effects on freshwater ecosystems (Patel & Sharma, 2023). Furthermore, the impact of emerging contaminants, such as microplastics present in wastewater, on freshwater biodiversity is a growing area of concern (Khan & Ahmed, 2025). Several substances are present in sewage, which can potentially impact seafood, plant and animal communities in different ways.  Common pathogens such as Vibrio spp., Salmonella, Shigella, Listeria monocytogenes, and Aeromonas spp. Are frequently Associated with seafood-related illnesses, particularly when products are consumed raw or Undercooked, or harvested from contaminated waters. The risk is exacerbated in environments with inadequate sanitation, improper handling, or cross-contamination during processing and Distribution.The presence of antimicrobial-resistant (AMR) strains further complicates treatment and control Efforts, making routine surveillance, stringent hygiene practices, and public education essential. This article therefore advocates for preventative strategies—such as monitoring harvest waters, enforcing seafood safety regulations. It also recommends that proper cooking is critical to reducing the incidence of foodborne infections and that continued Research and a One Health approach are needed to manage the evolving risks of bacterial Pathogens in seafood and safeguard public health.

Supercapacitors have emerged as promising energy storage devices owing to their high-power density, rapid charge-discharge capability, and long operational lifespan. However, optimizing their design and operating conditions remains a complex task due to the presence of multiple conflicting objectives, including energy density, power density, cost efficiency, and cycle life. This study proposes an artificial intelligence (AI)-driven multi-objective optimization framework that integrates evolutionary algorithms such as Genetic Algorithms (GA), Particle Swarm Optimization (PSO), and the Non-dominated Sorting Genetic Algorithm II (NSGA-II) with neural network-based surrogate modeling. The surrogate model is trained to approximate the nonlinear relationships between design parameters and performance metrics, thereby reducing computational and experimental burden during optimization. NSGA-II is employed to generate Pareto-optimal solutions that capture the trade-offs among competing objectives, enabling informed decision-making in material selection and structural configuration. The results demonstrate that the proposed AI-based framework significantly accelerates the design process, minimizes experimental costs, and enhances the efficiency of supercapacitor performance optimization. This approach provides a robust, scalable, and data-driven methodology for advanced energy storage system development.

This study evaluates Public Relations Strategies for Conflict Resolution in Nigerian Universities using a Case Study of Rivers State University. The objectives included identifying the specific public relations strategies employed by the university in addressing conflicts such as student fee protests, staff union disputes, and community-related tensions, and assessing their effectiveness in preventing, managing, and resolving conflicts among stakeholders. The research was anchored on three theoretical frameworks: the Excellence Theory of Public Relations, the Conflict Transformation Theory, and the Situational Crisis Communication Theory (SCCT). A mixed-method survey research design was adopted, combining quantitative and qualitative approaches for a comprehensive analysis. The total population of 30,462 comprised 12 Public Relations staff members, 2,450 academic and non-academic staff, and 28,000 students. Using Taro Yamane’s formula, a sample size of 350 respondents was determined. Data were collected through structured questionnaires and semi-structured interviews, validated by experts and tested for reliability. Quantitative data were analyzed using simple percentages and Weighted Mean Scores (WMS), while qualitative data were analyzed thematically. Findings revealed that Rivers State University employs a combination of dialogue, stakeholder meetings, community engagement, and internal communication channels to manage and resolve conflicts among students, staff, and external stakeholders. Public relations strategies such as timely communication, dialogue facilitation, and participatory decision-making were found effective in preventing and managing conflicts. The university has also introduced innovative PR initiatives such as stakeholder forums, participatory dialogue series, and trust-building workshops that foster mutual respect and institutional harmony. The study concluded that public relations play a vital role in fostering peace, understanding, and cooperation within the university community. It emphasized that effective PR practices have contributed significantly to building trust, promoting transparency, and preventing conflict escalation. It further suggested the enhancement of participatory communication, increased funding and autonomy for the PR department, and continuous capacity-building programmes for PR officers to strengthen conflict management effectiveness.

Further development of Artificial Intelligence (AI) and, especially, in such complex systems as Deep Learning (DL) and Large Language Models (LLM) resulted in their widespread application to the essential fields of human activity, such as healthcare, finance, and education. Non-linear, complex designs of these high-performance models, however, make them black boxes that are hard to understand, and their inner workings and decision-making are hard to interpret. This has brought up a high concern on the issue of trust, accountability, and ethical governance. This paper evaluates how the Explainable Artificial Intelligence (XAI) can alleviate this issue by rendering black-box models more transparent, understandable and interpretable to end-users and stakeholders. XAI solves the problem of interpreting complex algorithms and making them human friendly by offering ways of describing the processing of input data and decision formation. The significance of XAI is supported by the necessity to comply with the regulation, such as the General Data Protection Regulation (GDPR) that requires accountability of the automated decision-making, as well as the requirement to be fair, to detect and reduce biases hidden in the models. The three primary approaches of XAI methods are; Model-Agnostic Post-Hoc Interpreters (MAPHI): Techniques that are used after a model is trained, such as LIME and SHAP, that explain a prediction locally or globally; Intrinsically Interpretable Models (IIMs): Models that are inherently interpretable, such as decision trees though they can be less predictive power than LLMs; Overarching Frameworks and Auditing (OFA): Governance frameworks such as Responsible AI (RAI) that enact principles like Fairness, The problems of XAI, including the inherent trade-off between model accuracy and interpretability and the threat of explanation hacking are also addressed. To solve these problems, models such as OpenXAI are being studied to standardize the technical critique of the methods of explanation in terms of such important measures as faithfulness, stability, and fairness. Finally, XAI is not merely a technical requirement but an ethical foundation of successful AI implementation, as it is necessary to make the systems more human-centred and transparent, to allow building more trust and to enable the responsible AI development.

Invasive flora species (IFS) pose a significant threat to wetland biodiversity and the ecosystem services upon which local communities depend. This study investigated the ecological effects of invasive flora in the marshlands of Ikeduru, Imo State, Nigeria. The research mapped the spatial distribution of dominant invasive species, assessed their perceived impact on livelihood-critical ecosystem services, analyzed statistical relationships between invasive coverage and native biodiversity and soil/water quality, and developed a community-inclusive management framework. Data were collected via systematic field surveys using quadrats and transects, GPS mapping, laboratory analysis of soil and water samples, and a structured EEIFSESMI instrument administered to 396 households across 12 randomly selected communities. Results indicated a high prevalence of Nymphaea lotus (White Lotus) and Pistia stratiotes (Water Lettuce), with distinct aggregation patterns along disturbed edges. A significant negative correlation was found between invasive species coverage and native plant diversity indices (Shannon-Weiner, p

The sustenance of the global economy hinges on the ability of developing nations to maintain the stability of the banking systems in their countries. There are globally accepted Business Continuity and Disaster Recovery (BCDR) and planning systems, but in Nigeria, the effectiveness of such systems is greatly reduced in the light of the peculiar convergence of severe localized risks and systemic weaknesses. The present contribution, then, is the result of a systematic review involving ten selected journals and a strategic gap analysis, to develop a customized BCDR for the Nigerian Financial Services Sector. The review found the following three most important gaps: (1) Lack of specificity to the cyber (fraud) architecture in the sub-region which is particularly susceptible to financial fraud interdiction and advanced social-engineering scams. (2) Lack of coherent models for recovery with the Central Bank of Nigeria (CBN) and reputational losses, regulatory compliance, and other model recoveries interwoven. (3) A strong need to localize the metrics, such as Customer Accessibility, Recovery Time Objective, and other indicators of success, which are global to infrastructure realities (e.g., grid instability, fiber vulnerability). The suggested Strategic Roadmap is built on three focused pillars (Secure, Adapt, and Measure), aiming at assisting financial entities to obtain a resilience framework that incorporates technology with sustainability. The framework focuses on delivering more than just uptime; it aims to provide uninterrupted customer access, regulatory provision, and operational trust. It will facilitate advanced work on resilience computational models with multiple recovery levels, predictive risk models, and other AI-based perturbation models meant for developing economies.

The study was motivated by the persistent limited participation of community members in development discussions, where silence often replaces active engagement during decision-making processes. The objectives were to identify the forms of silence, examine the factors responsible for silence, and assess its effects on community participation and development processes in Elele. The study adopted the survey research design. The population of the study comprised the projected population of 250,000 residents of Elele community, while a sample of 380 respondents was selected for the study. Data were collected using a structured questionnaire and analyzed using the Weighted Mean Score (WMS) based on a four-point Likert scale. Findings revealed that silence in Elele manifests mainly as fear of speaking, respect for authority, and avoidance of conflict during development meetings. It was also found that silence is influenced by intimidation, cultural norms, lack of confidence, and hierarchical structures within the community. Furthermore, the study showed that silence negatively affects community development by limiting participation, weakening feedback mechanisms, distorting decision-making, and contributing to poor implementation of development projects. The study concluded that silence is a significant communication phenomenon that shapes participation and outcomes in development communication processes in Elele community.

This study geotechnically characterized reclaimed wetlands soils for building bearing capacity in Obio-Akpor Local Government Area, Rivers State. The study adopted a true experimental research design. Twelve soil samples were collected from six reclaimed wetlands in the study area. The laboratory results obtained from the soil samples were analyzed using mean and charts to answer the research questions, while Analysis of Variance (ANOVA) was employed to test the hypotheses at the 0.05 level of significance. The study found spatial variability in soil properties range: temperature (28.90- 30.53°C), pH (7.60-8.42), EC (1,428.67-1,760.50 µS/cm), and organic carbon (1.46-2.17%). Bulk density ranged from 0.96-1.79 g/cm³, with Rukpokwu lowest and Rumuola highest, indicating variations that may affect foundation stability, bearing capacity, and drainage, and performance. Also, the significant differences (p= .026) in soil parameters, and highly significant (p= .001) bulk-density variation, suggests spatial heterogeneity that may affect soil stability and bearing capacity in the study area. The study concludes that spatial variability in soil properties and bulk density may influence foundation stability, and building performance, necessitating sitespecific geotechnical assessment in Obio-Akpor reclaimed wetlands. The study recommended that the development-control authorities in Rivers State and Obio-Akpor local government, should require sitespecific bulk-density assessment and professional foundation recommendations from qualified geotechnical engineers, civil/structural engineers, and certified soil-testing laboratories before approving low-, medium-, high-rise, and specialized structures on reclaimed wetlands. Alongside, relevant planning authorities should mandate physicochemical soil testing before construction in Rukpokwu, Eligbolo, Eliozu, Minikpeti, Rumunduru, and Rumuola, variations in soil softness, texture, pH, electrical conductivity, and organic-carbon content may affect foundation performance.