Books
Permanent URI for this communityhttp://197.211.34.35:4000/handle/123456789/3
Books
Browse
5 results
Search Results
Item Food safety forecasting using internet of things and machine learning(ABU-Zaria, 2023) Fadeyi John Oluwole; A. U. Usman; Oyewobi S. StephenFood sector is a significant part of the economy but it faces challenges with food spoilage, especially in meat, fruits, and vegetables. This issue involves food items, especially meat, fruits, and vegetables, going stale and often reaching consumers unnoticed. Additionally, during the food chain there may be instances where the food may still be within the proposed shelf life but may be spoilt before it gets to the consumer, therefore, it is important to test them and envisage when it will be inedible. This paper presents a predictive model which is used to forecast when the fruits will he inedible via the use of time series data generated from internet of things (loT) based device. The loT device developed in this research is used to monitor the decline of the freshness of the fruit to the state of inedibility. This device measures parameters such as alcohol, and ammonia around the fruit, as such large amounts of real-time data are generated. A web server is used for the storage of data values sensed in real time and also for the analysis of results. Long Short-Term Memory (LSTM) predictive model is used to forecast the time the fruit will be inedible via the use of time series data harvested from the cloud. The implementation of this technology enhances traceability. minimizes food wastage, and, most importantly, protects consumers from foodborne illnesses. Keywords: Food safety, loT, Machine learningItem Fuzzy Logic-Based Electrical Energy Management of Building(international engineering conference (IEC 2023), 2023) Abdul-Azeez Dauda; Oyewobi S. Stephen; Farouq E Shaibu Umar S. DaudaThe modern trend in electrical energy distribution is the soft island system. Manual and conventional procedures struggle when dealing with multiple sources, loads, and constraints because they need intelligence to perform at their best. This work, therefore, proposes an intelligent fuzzy logic-based controller for building energy management system that controls solar panel and inverter, grid, battery and inverter, and generator in a way that prioritizes the use of the cleanest and most affordable source. To simulate the developed system in MATLAB Simulink, real data from offices in the engineering complex of Niger State Polytechnic Zungeru, Nigeria, was used. Results showed that when compared to the conventional distribution board, the developed algorithm conserved power by 33.71%.Item Dipole Antenna Design Leveraging Optimization Techniques.(Harvard International Journal of Engineering Research and Technology, 2023-12-31) Akamike, Ogechi; Caroline Alenoghena; Oyewobi S. StephenDipole antennas are the commonest and simplest type of antennas in terms of design. Antenna Engineers and researchers have been looking for ways to design dipole antenna that can address specific needs of their clients, without incurring additional expenses on the available resources. Therefore, optimization techniques becomes necessary to be leveraged in order to design cost-effective antennas. This paper presents dipole antenna and its varieties as well as different optimization techniques that can be leveraged to design dipole Omnidirectional antennas that can stand the test of time. Most dipole antenna designs carried out by some researchers did not include integrating optimization techniques in their approach.Item Development of an artificial intelligent agent for library hard materials distribution operations(AFIT Journal of Science and Engineering Research, 2024) Kufre Esenewo Jack; Waheed Moses Audu; Lanre Joseph Olatomiwa; Umar Suleiman Dauda; Bello Kontagora Nuhu; Oyewobi S. StephenIn recent years, automation has emerged as a technological solution that is gaining grounds across various fields. The task of managing updates like sorting, shelving and documenting large collections of books in the shortest possible time is stressful with limited personnel and can be automated. Other important library tasks such as consultation can, therefore, be undertaken effortlessly. This paper which explores the use of automated solutions for the purpose of book sorting in libraries, proposes the You Only Look Once version 5 (YOLOv5) model to detect books, then performs optical character recognition using EasyOCR on the detected book. After the characters have been extracted, the system then classifies the book into its respective library section using OpenAI. The result from testing the system shows that the book detection model gave an accuracy of 74%, the EasyOCR performed with an accuracy of 91% with variations across different image formats. For simplicity, images used were stored in jpg formats for faster execution time and easy processing. The paper emphasizes the revolutionary impact of AI and machine vision in educational institutions, especially in libraries beyond what can be imagined, pushing for the formation of library robots. This project's benefits include object detection and intelligent book categorization which phase to a new direction for study and advances library automation technologies that boost education.Item Evaluating the effectiveness of machine learning models for path loss prediction at 3.5GHz with focus on feature extraction(Nigerian Journal of Technology, 2024) F. E. Shaibu; E. N. Onwuka; N. Salawu; Oyewobi S. StephenAccurate path loss prediction is vital for efficient resource allocation, interference reduction, and overall network reliability in 5G networks, particularly in the widely deployed mid-band frequency spectrum (such as 3.5 GHz). This study evaluates the effectiveness of machine learning models for path loss prediction at 3.5 GHz with a focus on feature prioritization. A feature selection method, recursive feature elimination, was used to identify significant features from datasets obtained through measurement campaigns, weather stations, 3-D ray tracing, geographical data, and simulations. Out of eighteen features, eleven, including new environmental features, were identified as significant features contributing to path loss. These selected variables were then utilized to optimize and train four common machine learning models (ANN, XGBoost, RF, and k-NN) to evaluate their performance in predicting path loss in a specific urban area called an irregular urban environment. The performance of these models was assessed by comparing their predictions with the measured path loss. The Random Forest model closely matched the measured path loss over the entire path length in both LoS and NLoS scenarios, achieving the lowest MAE of 0.15 dB and RMSE of 0.57 dB in the LoS scenario and 0.62 dB and 1.42 dB in the NLoS scenario, with R2 scores of 0.999995437 and 0.999996828, respectively. This indicates its superior performance in predicting path loss in the urban environment