Henry Onyebuchukwu Ordu
Unknown Affiliation

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

Optimized Vessel Scheduling Model Using Multilayer Perceptron Algorithm Henry Onyebuchukwu Ordu; Joseph Tochukwu Odemenem
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 4 No. 3 (2025): September 2025
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v4i3.6031

Abstract

Efficient vessel scheduling is crucial to the performance and profitability of maritime terminals, yet conventional approaches often struggle to accommodate the dynamic, nonlinear interactions among vessel arrivals, cargo handling requirements, and berth availability. This study presents a Multilayer Perceptron (MLP)–based scheduling framework that models these complex relationships and delivers actionable berth assignments in real time. Leveraging an integrated dataset of historical arrival and departure timestamps, cargo throughput, and occupancy records, the MLP model was trained on 80% of the data and rigorously tested on the remaining 20% Performance was assessed using metrics such as vessel turnaround time, berth utilization rate, and scheduling accuracy. Experimental results reveal that our MLP-driven scheduler achieves a 15% reduction in average turnaround time and a 12% increase in berth utilization. Remarkably, the neural network maintains high levels of schedule adherence even under peak-demand scenarios, minimizing idle berth time and streamlining cargo flow. These findings underscore the adaptability of advanced machine learning techniques to the evolving demands of port operations.
A Convolutional Neural Network-Based Real-Time Behavioral Detection System for Preventing Cheating in Online Examinations Muktar Abubakar Muhammed; Henry Onyebuchukwu Ordu
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 5 No. 1 (2026): Maret 2026
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v5i1.7676

Abstract

The integrity of online examinations has become a growing concern in digital education, particularly following the rapid shift to remote learning. This study presents the development of a Convolutional Neural Network (CNN)-based Real-Time Behavioral Detection System and Prevention of cheating in online examinations. Specifically, the study identifies and classifies common visual behaviors associated with cheating, such as frequent eye movement, head turning, and the presence of unauthorized individuals. A CNN model was designed and trained on a curated dataset of annotated behavioral frames. The model achieved a classification accuracy of 91.7%, precision of 89.5%, recall of 92.3%, and an F1-score of 90.9%, demonstrating strong performance in real-time cheating behavior detection. A working prototype was developed using Python, TensorFlow, and OpenCV, and successfully integrated into a live monitoring interface capable of issuing alerts, logging incidents, and generating post-exam reports. The system's performance was evaluated across various test scenarios, showing consistent results with an average latency of 0.72 seconds per frame, making it suitable for real-time deployment.. Its implementation offers significant value to educational institutions, exam regulators, and EdTech platforms seeking to ensure fairness and trust in digital examinations.
Development of an Enhanced Predictive Model for Road Accident Occurrence in Nigeria Chukwudi Ugbaja; Friday E. Onuodu; Henry Onyebuchukwu Ordu; Emmanuel J. Izionworu
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 5 No. 1 (2026): Maret 2026
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v5i1.7700

Abstract

Road accidents in Nigeria rank as the second highest globally, with 33.7% of deaths per 100,000 persons occurring annually. This study developed and tested a predictive model for road accident occurrence using Artificial Neural Networks (ANN) to address the technological gap in Nigeria's road safety management systems. A feed-forward neural network architecture comprising 52 input neurons, three hidden layers (32, 16, and 8 neurons) with ReLU activation, and a single sigmoid output neuron was designed. Dropout (0.3, 0.3, 0.2) and L2 regularization (0.001, 0.001, 0.0005) were incorporated to address sample size constraints. The dataset comprised 2,847 records from FRSC, NEMA, and NBS (2018-2023) across twelve Nigerian states, with 24 features spanning road, environmental, driver, and vehicle factors. Stratified random splitting yielded 1,994 training, 570 validation, and 283 temporally distinct test records. The model achieved 84.5% accuracy (95% CI: 79.8%-88.5%), 77.0% recall, 89.4% specificity, and 0.89 AUC on independent test data—a 13.5 percentage point improvement over the existing K-modes system (p<0.0001). Five-fold cross-validation confirmed stability (84.3%±0.6%). Feature importance analysis identified speeding (18.4%), alcohol impairment (15.2%), wet roads (11.8%), night driving (9.4%), and lane discipline (8.1%) as dominant predictors, with human factors accounting for 45.3% of predictive power. This study provides the first evidence-validated ANN-based accident prediction model calibrated for Nigeria, establishing a reproducible methodological template for developing contextually-adapted predictive systems in data-constrained environments while demonstrating statistically significant and practically meaningful improvement over existing approaches.
Development of a Content Creation Model Using Natural Language Generation Paul-Odeli Jonathan Ateko; Constance I. Amannah; Henry Onyebuchukwu Ordu; Izionworu, Emmanuel J.
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 5 No. 1 (2026): Maret 2026
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/jomlai.v5i1.7726

Abstract

The increasing demand for scalable, high-quality digital content has exposed the limitations of manual content creation and existing Natural Language Generation (NLG) systems, particularly in terms of domain specificity, ethical reliability, and readiness for optimization. This study addresses this gap by developing NLG-ACCO, a transformer-based model for automated content creation and optimization in educational, media, and digital marketing applications. Transformer-XL was selected over newer architectures like Llama-3 or Mistral because it models longer contextual dependencies beyond fixed-length segments—essential for coherent paragraph-level content—while offering a better trade-off between performance, computational efficiency, and transparency under resource-constrained conditions. The model integrates domain-aware fine-tuning, reinforcement learning, SEO optimization, and ethical safeguards, including bias detection and factual verification. Evaluation used BLEU, ROUGE, readability indices, and Perplexity. NLG-ACCO achieved a BLEU score of 0.79 (baseline: 0.61) and ROUGE-L of 0.76 (baseline: 0.36). Perplexity dropped from 45.2 to 27.8, indicating more coherent predictions. Readability improved by 24%, post-editing time decreased by 38.5%, and bias detection mitigated 87% of flagged cases. These results demonstrate that integrating optimization and ethical controls within Transformer-XL frameworks significantly enhances content quality and reliability.