Covenant University

Covenant University Repository
Not a member yet
    15288 research outputs found

    A NOVEL ALGORITHM FOR IMPROVED TRACTABILITY AND EXPLAINABILITY OF INFERENCES ON HIGH-DIMENSIONAL DATA USING CONDITIONED ATTENTION MECHANISM

    Get PDF
    The availability of a large amount of data that can be mined is a challenge for data mining techniques that cannot keep up with the exponentially increasing quantity of data. Most of the classical techniques have been reported inefficient to handle large amount of data. The aim of this work is to develop a novel attention mechanism for high-dimensional tabular data understanding for knowledge representation that can enhance tractability of decision rule learning. The conditioning theory was mathematically formalized as an attention mechanism that models an unconditioned response (or reflex response) to unconditioned stimulus by pairing conditioned stimulus with unconditioned stimulus. The formalization of the conditioning theory built was translated into an algorithm presented as conditioned attention mechanism and applied on input tabular dataset producing vector embeddings of attribute-value pairs. The learned representation of attribute-value pairs as vector embeddings are then exploited to build decision rules using a developed novel Coupling-Sorting-Branching technique (CoSoBra), formalized as energy-based model optimization. A Python-based compression algorithm was developed to reduce the dimension of the resulting very large vector embeddings while maintaining the quality of the attribute-value pairs representations. The developed novel Conditioned Attention High Dimensional Rule Learning (CAHDRL) algorithm, which captures both CoSoBra algorithm and Condition Attention mechanism was benchmarked against existing approaches for decision rule learning on ten datasets from six different domains. The efficacy of the CAHDRL approach was evaluated with respect to the effectiveness of the extracted decision rules and its efficiency during data processing. An ablation study was carried out to assess the efficiency CAHDRL. Results of the evaluation on the 8 first datasets showed that CADHRL had higher accuracy of 0.85 in average against 0.8175 in average for the EAV-60 model, equal length of decision rule of 2.125 in average against 2.125 in average for the reduct-alpha-0.1 model and a better support of 392.5263 in average against 350.2663 in average for the reduct-alpha-0.1. t-distributed Stochastic Neighborhood Embedding (t-SNE) was used to visualize the embeddings on a 2D manifold. The results support literature findings on Mushroom dataset that “odor=creosote” pair is a good descriptor of the poisonous mushroom. The second visualization results using “Adult” dataset showed that USA natives are more likely to earn $50,000.00 and above. Wilcoxon test of significance of efficacy results for CAHDRL validated its relatively superior performance against existing approaches, yielding values of p-values inferior to 0.05 for accuracy (3 of the seven comparisons), length of decision rules (6 of the seven comparisons) and support (4 of the five comparisons) respectively. CAHDRL can effectively find embedding vectors of attribute-value pairs and efficiently extract decision rules from diverse datasets for classification tasks, with a higher tractability and inherent explainability. CAHDRL also demonstrates a great scalability feature that allows for the efficient processing very high dimensional datasets by the parallelizable nature of CAHDRL

    PREVALENCE OF HISTIDINE-RICH PROTEIN II GENE DELETION IN PLASMODIUM FALCIPARUM AMONG SUBJECTS IN ADO-OTA, OGUN STATE.

    Get PDF
    The key to reducing malaria-related deaths in areas with high transmission rates is prompt diagnosis and appropriate treatment. Malaria Rapid diagnostic tests (RDTs or mRDT) that utilize histidine-rich protein 2 (PfHRP2) as a biomarker are essential for quickly and accurately diagnosing P. falciparum infections in endemic regions like Nigeria. However, PfHRP2 gene deletion and the variability of amino acids threaten the mRDT test sensitivity. This study was carried out to investigate the prevalence of HRP2 gene deletion in Plasmodium falciparum among symptomatic subjects in Ota. Ethical clearance from the Covenant Health Ethics Review Committee (CHREC) was obtained. Three-hundred and ninety-six samples were collected from subjects visiting four healthcare facilities in Ota, Ogun State, Nigeria and tested for malaria using HRP2-based RDT. The presence of parasites in RDT false negative cases and microscopy positive cases were validated using nested Polymerase Chain Reaction (PCR). Thereafter, exon 2 of PfHRP2 was amplified, and Sanger sequenced. The prevalence of malaria was recorded at 57.07%, with the age group below 5 years showing the highest prevalence of 49.15%. Out of 50 samples, forty-seven were positive by 18sRNA quantitative PCR. Single-copy gene MSP1 showed an overall multiplicity of infection of 1.5. The MAD20 allele had the highest frequency (72.72%), the K1 allele (54.45%) and the RO33 allele recorded the lowest frequency (27.27%) among the samples that amplified the MSP1. The prevalence of HRP2 gene deletion was recorded at 15.4% (2/13) among samples with amplified single-copy genes using Nested PCR. More studies using larger sample sizes for genotyping and more sensitive techniques like digital droplet PCR are recommended to determine the full extent of PfHRP2 gene deletion in Nigeria

    Assessment of Hydropower Generation and Green Hydrogen Production Potential in Jebba Dam, Nigeria, West Africa

    Get PDF
    Hydropower can play a significant role in advancing the production of green hydrogen. However, hydro-climatic variability impacts hydropower production and thus the hydrogen potential. In this study, we developed novel analytics to understand how hydropower inter-annual variability translates to hydrogen production. We address this by; (i) analyzing Jebba dam hydro-climatic variables and associated hydropower generation (ii) translating the annual and quarterly hydropower production into hydrogen using five assumed scenarios; (iii) estimating the reelectrification potential and (iv) determining the quantity of petrol (or gasoline) that would be replaced and the amount of CO2 and CO that would be avoided. We find that hydropower energy generation has increased significantly at the station. The estimated annual and quarterly green hydrogen potentials indicated that the highest potentials were 59,111 tons and 18,744 tons and have a re-electrification potential of 1,182 GWh and 374 GWh, which can replace 0.224 million liters and 0.071 million liters of petrol (or gasoline) in the year 2021 and the fourth quarter of 2021–04, respectively. This would prevent 0.52 million kg of CO2, 0.92 thousand kg of CO in the year 2021, and 0.163 million kg of CO2, 0.293 thousand kg of CO emissions in the fourth quarter of 2021–04. The study concludes that the impact of hydro-climatic variation on hydropower generation affects green hydrogen production potential. Nevertheless, using a percentage of hydropower energy can present a unique opportunity to move the nation toward the production of green hydrogen energy as a long-storage solution for rural areas' re-electrification and to meet electricity demand when the hydropower dam’s water storage is low. Furthermore, the adoption of a green hydrogen energy solution can contribute to the nation's and global climate change mitigation efforts

    Predicting Cross-Selling Health Insurance Products Using Machine-Learning Techniques

    Get PDF
    This study delves into the utilization of Machine Learning (ML) techniques for predicting health insurance cross-selling behavior in South African consumers. The main goal is to create a robust ML model that assists health insurance companies in pinpointing potential customers with higher probabilities of purchasing additional health insurance products. Employing quantitative methodology, the study extracted consumer data and applied various ML algorithms such as random forest, K-nearest neighbors, XGBoost classifier, and logistic regression using Python. Tailored feature engineering techniques were employed to enhance predictive accuracy. Analyzing 1,000,000 customer records with 16 features, Random Forest emerged as the topperforming model, achieving an accuracy score of 0.99 and F1 score of 1.00. The study reveals that customers aged 25–70, with prior insurance and longer service history, are more inclined to purchase additional health insurance products. These findings provide actionable insights for refining marketing strategies, boosting customer acquisition, and increasing revenue

    I-V response test of 60–150W mono-crystalline solar panel

    Get PDF
    This work investigates the discrepancies in electrical parameters of monocrystalline solar panels between Ago-Iwoye weather conditions and the manufacturer’s specified ideal conditions. Manufacturer’s specifications are typically based on 1,000 W/m2 global solar irradiance, AM 1.5, and 25°C operating temperature, while actual weather conditions at installation sites can vary significantly. Mono-crystalline (single-crystal) silicon solar panels of capacities 60, 80, 100, and 150W were evaluated through current-voltage (I-V) response tests at an installation site in Ago-Iwoye, Nigeria, with solar irradiance exposure from 11 a.m. to 3 p.m. The analysis of I-V and P-V curves revealed a significant reduction in maximum power output by 28.6%, 25.9%, 28.9%, and 19.36%, respectively, compared to the manufacturer’s stated values. This deviation underscores the importance of considering local weather conditions during solar PV projects, and we recommend adding an additional 20%–30% of the total solar panel capacity during installations to account for variations in solar irradiance and operating temperatures, ensuring optimal performance and effective solar power generation in Ago-Iwoye and similar areas

    Promoting the use of bioenergy in developing nations: a CDM route to sustainable development

    Get PDF
    The study focused on a forensic assessment of bioenergy utilization in developing countries, emphasizing how to improve bioenergy resources for a sustainable economy and development. First, a holistic review of energy sources and their impacts on sustainable development was highlighted. The study also looked at the relationships between various energy sources, both the supply and demand of energy, as the population statistics in developing countries. However, renewable energy is yet to be harnessed fully as it lagged by 0.7%. Shockingly, the African continent is backward in both the global energy supply and in tapping renewable energy sources, as only 6% is contributed. More so, applying a clean development mechanism (CDM) for energy utilization and control shows that net emission reduction and global temperature reduction of 1.5°C by 2030 is attainable. Thus, helping to realize low-cost and sustainable energy (SDG 7) and sustainable cities and communities (SDG11). Further, the assessment of the level of bioenergy availability and its utilization shows that 10% of the world’s energy is provided by bioenergy. Interestingly, it is available in every region in the form of biomass. The study is helpful for stakeholders, policymakers, and governments managing the affairs of the energy sector on the need to formulate policy that will guide effective bioenergy utilization and reduction in greenhouse gas emissions, especially in developing countries thereby, promoting sustainable economic growth

    STREET CULTURE AND CHARACTER DEVELOPMENT IN SELECTED HOLLYWOOD AND NOLLYWOOD FILMS

    Get PDF
    In contemporary society, the problem of street culture has been a recurring issue. Some Children or adults leave their homes to live in harsh conditions on the street, thereby imbibing behaviours which qualify as street culture. Several studies have examined street culture from the perspective of gangsterism, violence and music as represented in written texts with little attention paid to films. This study investigated street culture in Nollywood and Hollywood films and its impacts on the character of the actors and actresses in the selected films with a view to highlighting its implications for contemporary society. The study adopted the content analytical method in interrogating the selected films. A purposive sampling technique was used to select a total of twelve films comprising six Nollywood and six Hollywood films. The films were subjected to content and qualitative analysis to explore the elements of street culture in them. The study adopted Erikson’s Psychosocial Development Theory and Sarris’ Auteur Theory as the framework for analysis. Findings revealed that street urchins are victims of poor parenting, economic hardships, poverty, peer pressure, absence of a father figure, poor societal values, the death of a parent, neglect, abandonment, physical and emotional abuse and violence as portrayed in the selected films. The study found that the consequences of these factors on street urchins include maladaptive behaviours, indiscriminate sexual practice, substance abuse, prostitution, stealing, and pickpocketing. Further findings revealed that filmmakers deployed different narrative patterns to elucidate how the characters develop in the films. In addition, young people who engage in street vices in Hollywood and Nollywood films are labelled differently in the films. In Hollywood, they are referred to as ‘homeless youths’ or ‘homeless children’, but in Nollywood, they are called ‘street children’ or ‘children in the street’. It was revealed that in both Hollywood and Nollywood films, the absence of a father figure in a child’s life is detrimental to the character development of the child. Another major finding is that the plot structure revealed a pattern in which some of the characters in the Nollywood and Hollywood films went unpunished and prospered from their corrupt practices. This has implications for the propagation of values in any society. The study concluded that the influence of street culture made the characters engage in maladaptive behaviours due to the failure of caregivers to fulfil their responsibilities towards them. The study, therefore, recommended that parents should not shirk their duties towards their children and should provide enabling environments through which the children can feel safe and repose confidence in them. Also, every adult should help train children in their parents' absence. The Government should censor the quality of films produced to make them avenues for correcting societal ills and promoting right values in adults and children

    The Role of Nanotechnology in Proton Exchange Membrane Fuel Cell and Microbial Fuel Cell: The Insight of Nanohybrid

    Get PDF
    The utilization of proton exchange membrane (PEM) or polymer electrolyte membrane fuel cell (PEMFC) and microbial fuel cell (MFC) is attaining prominence for the production of energy; however, these technologies come with their challenges like the cost of platinum catalyst, which is fifty percent of the total budget of the technologies. Here, the role of nanotechnology in proton exchange membrane and microbial fuel cells is reviewed. Remarkable exploits have been committed to the use of nano-materials to synthesize catalyst that are earth-abundant as an alternative to -platinum catalyst. The design of these nano-materials must be intentional for the attainment of very large surface areas. In the quest of making fuel cells affordable, researchers have focused their attention on designing nanomaterials that will make fuel cells cost-effective. The outstanding chemical characteristics in addition to the physical characteristics of these nano-materials have resulted in all-encompassing investigations targeted at enhancing the performance and cost-effectiveness of fuel cells. The application of nano-materials has also been extended in areas such as PEMFC and MFC, by incorporating the nanomaterials in the membranes and electrodes for the purpose of enhancing the performance of the fuel cells. The future prospects of proton exchange membrane and microbial fuel cells in the light of nanohybrids discussed showed that nanohybrids yields outstanding mechanical properties together with chemical steadiness; increased selectivity and proton conductivity; increased electron transfer efficiency and the highest capacitance; increased output power density and the transference of charge; and upsurge kinetics of the electrodes which will enhance power production

    ADDRESSING THE LETHARGY IN THE CAPTURE FISHERIES INDUSTRY: ADVOCACY FOR PROGRESSIVE FISHERIES MANAGEMENT POLICIES

    Get PDF
    This paper examines the governance of capture fisheries industry in Nigeria with a view towards advocating for a progressive fishery management policies for the sector. Although there are extant policies in the fisheries industry (comprising legislative and administrative policies), offenders commit infractions with impunity, artisanal fishers are left without critical support while relevant agencies are unable to implement the critical aspects of the administrative policies that are geared towards boosting the fishing activities in the sector. There is a huge regulatory gap where many of the fishers are able to promote their mercantile instinct over the values embedded in legislation whereas policies meant to sustain small-scale fishers are hardly implemented. Therefore, in view of the huge regulatory gap, this paper identifies, reviews and interrogates some of the contents of the policies. It finds that aspects of the existing legal regime are outdated or unenforced, critical support that artisanal farmers need are withheld from them, and many of the administrative policies are hardly implemented. Therefore, in order to overcome the current comatose state of the fisheries sector, the paper suggests a progressive approach to the institution of fisheries management policies for the benefit of stakeholders including the artisanal fishers, relevant government agencies, citizens and the country

    Review of Aluminium Beverage Can Production and its Associated Problems

    Get PDF
    Aluminium beverage can production has been associated with many problems such as formation of necking wrinkle, bottom can, being blackened, strips on the inner can and the body printing. These problems cause severe reduction in the mechanical properties of the can and subsequent issues during the packaging of the food and beverages. Thus, this study seeks to look into the properties and types of aluminum and the probable causes of the problem during the production of aluminium. Findings revealed that enhanced valorization process is needful in attaining great quality of cans for a sustainable production

    13,900

    full texts

    15,288

    metadata records
    Updated in last 30 days.
    Covenant University Repository is based in Nigeria
    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! 👇