Department of Physics and Computer Science
Permanent URI for this collectionhttps://dspace.njala.edu.sl/handle/20.500.144402/139
Browse
5 results
Search Results
Item A review of artificial intelligence of things(Institution of Engineering and Technology, 2022-11-23) Conteh, A.The Internet of Things (IoT) has introduced a current and up to date global powered with the aid of networking producing diverse heterogeneous records for a very long time in so many fields and applications. In removing entire facts from those records, advance artificial intelligence (AI) technology, particularly deep learning (DL), has shown a success in facilitating records analytics, destiny prediction, and choice making. The combination of AI and the IoT has appreciably promoted the fast improvement of Artificial Intelligence-of-Things (AIoT) structures that examine and reply to outside stimuli extra intelligently without human intervention. However, it's far difficult or infeasible to method huge quantities of records in the cloud because of the unfavorable effect of records volume, velocity, and veracity and deadly transmission latency on communication infrastructures. This article conducts a survey of the internet of things and artificial intelligence orchestrated architecture for flexible artificial intelligence of things (AIoT) systems. It begins by explaining fundamental concept of the internet of things and artificial intelligence with this it explains the overall artificial intelligence of things architecture show and example of AIoT to illustrate how AI can be used in real world and summarize IoT and AI application.Item Assessing the Impact of E-Learning Adoption on Learning Behavior in Sierra Leonean Higher Education(Sciencedomain International, 2024-03-23) Sawaneh, Ibrahim AbdulaiThis study aims to analyze the adoption of e-learning in higher education institutions in Sierra Leone and its impact on learning Behavior using the Technology Acceptance Model (TAM). The research employs a survey design, utilizing Structural Equation Modeling (SEM), to analyze the responses of students and lecturers. The study was conducted in Sierra Leone, involving ten universities. The duration of the study is not specified. 728 students and 327 lecturers were surveyed to assess their attitudes towards e-learning. The TAM was used to predict behaviors based on the perceived usefulness and ease of use of e-learning tools, such as mobile devices and social networking applications. The findings indicate that e-learning is more likely to be adopted if the tools are perceived as useful and easy to use. Additionally, prior experience with computers enhances the positive perception of e-learning's attributes and attitudes toward its utilization. The study highlights the potential of e-learning in developing countries like Sierra Leone. It provides insights for stakeholders in the education sector to develop strategies that can promote the adoption of e-learning in regions with similar developmental profiles.Item DeepEVD: Integrating Epidemiological data into deep learning frameworks based on spatio-temporal feature learning for EVD forecasting(Elsevier BV, 2025-08-01) Fofanah, Abdul JosephThe paper introduces DeepEVD, an innovative framework that integrates human mobility data to forecast Ebola Virus Disease (EVD) outbreaks. Traditional epidemiological models often struggle to account for the dynamic nature of human movement, which is crucial for understanding EVD transmission. DeepEVD leverages diverse mobility data sources, including phone records, GPS traces, and social media posts, to extract significant spatio-temporal features. It utilises Graph Convolutional Networks (GCN) and Long Short Term Memory (LSTM) networks to establish connections between mobility patterns and EVD cases across both space and time. The framework was tested on real-world datasets from the 2014-2016 West Africa outbreak and the 2015-2016 Sierra Leone outbreak, demonstrating a 5%-10% reduction in forecasting errors compared to baseline methods. Ablation studies reveal the impact of various data sources and feature extraction methods on accuracy. DeepEVD not only delivers state-of-the-art performance, but it also provides actionable insights for EVD prevention and control. Implementation of the proposed DeepEVD can be accessed here https://github.com/afofanah/DeepEVDMob.Item Assessment of Background Noise Levels in Four Commercial Areas of Freetown, Sierra Leone(2025-06-04) Abu, Mohamed MustaphaEnvironmental noise is a significant public health issue, particularly in rapidly urbanizing cities where commercial activities contribute to elevated noise pollution. Despite its numerous implications, few studies have focused on noise levels in commercial environments in developing countries. This study aims to fill this existing gap by assessing background noise levels in four commercial areas of Freetown, Sierra Leone: Aberdeen Market, Godrich Street, Jui Junction, and Sani Abacha Street. Using a calibrated handheld sound level meter, measurements were taken during daytime, evening, and nighttime periods over one week. Noise levels were analyzed using the day-evening-night equivalent continuous noise level (Lden) and statistical methods to assess compliance with World Health Organization (WHO) guidelines. Results revealed that all areas exhibit average noise levels above the WHO recommended limit of 70 dBA during daytime and evening hours, with Godrich Street and Aberdeen Market exhibiting the highest average daytime noise level (95.35 dBA) and the lowest average nighttime noise level (53.05 dBA) respectively. Godrich Street recorded the highest Lden value and shortest allowed exposure of 93.34 dBA and 1.17 hours respectively, indicating the most significant risk of prolonged exposure. Recommendations include creating designated quiet zones tailored to local commercial layouts, stricter enforcement of noise regulations, public awareness campaigns on hearing protection, and regular noise monitoring to guide urban planning. This study provides crucial insights for mitigating noise pollution and improving urban sustainability in developing nations.Item Integrated neutral-hydrolysis recycling of mixed poly(ethylene terephthalate) and poly(butylene terephthalate) waste: System-level design, techno-economic, and environmental evaluation(Elsevier BV, 2026-06-30) Boré, AbdoulayeMixed poly(ethylene terephthalate)/poly(butylene terephthalate) (PET/PBT) waste represents an increasing fraction of heterogeneous polyester streams with limited compatibility with mechanical recycling within municipal material recovery facilities. This study evaluates a continuous neutral-hydrolysis process for a nominal 70/30 PET/PBT feed using Aspen Plus simulation, techno-economic analysis, and cradle-to-gate life-cycle assessment (LCA). The framework integrates literature-based reaction kinetics with simulated separation, economic, and environmental inventories to assess the system-level implications of mixed PET/PBT neutral hydrolysis. At 270 °C, 60 bar, and 30 min residence time, the simulation predicted 80.75 kg/h terephthalic acid (TPA) in the recovered solid cake from 100 kg/h mixed PET/PBT feed, corresponding to 97.1 % of the stoichiometric maximum. The recovered cake contained 96.45 wt% TPA prior to product polishing. Thermal demand is concentrated in hydrolysis and downstream separation, with the base-case simulation used to quantify external utility requirements. A separate pinch-based utility-displacement analysis indicates that recovering roughly 40 % of the available hot-cold overlap could reduce external utility demand; this value represents a thermodynamic integration target, not demonstrated heat-exchanger-network performance. With glycol coproduct credits included, the estimated minimum selling price was \$3.66/kg TPA, while LCA estimated a greenhouse-gas intensity of 2.68 kg CO 2 -eq/kg TPA. Within the modeled system boundary, neutral hydrolysis may complement mechanical recycling for PET/PBT-rich waste streams. Pilot-scale validation with representative mixed PET/PBT feedstocks and assessment of contaminant partitioning, solids handling, and recycle-water stability remain essential next steps.