Papers by Ekemini Thompson
virgilAI, 2025
Large Language Models (LLMs) have revolutionized natural language processing, but their generalis... more Large Language Models (LLMs) have revolutionized natural language processing, but their generalist nature often leads to suboptimal performance in domain-specific tasks like aviation question-answering (Q&A), where precision in regulations (e.g., FAA standards) and safety are paramount. This paper presents a comprehensive framework for fine-tuning Meta-Llama-3-8B-Instruct on the AviationQA dataset—a synthetic corpus of 1 million aviation-focused Q&A pairs—to create an Aviation Q&A Assistant. Employing Low-Rank Adaptation (LoRA) for efficient parameter tuning, we achieve a ROUGE-L factuality score of 0.502 and a 100% safety refusal rate on adversarial queries, outperforming baseline Llama by 25% in domain coherence. Our methodology integrates Supervised Fine-Tuning (SFT) with system prompt engineering for alignment, addressing challenges like dataset noise and hallucination. Evaluations encompass factuality (ROUGE-L), safety (refusal rate), and bias (HELM-lite), demonstrating robust performance for pilot training applications. Deployed via Hugging Face Spaces, this MVP serves as a scalable blueprint for domain-adaptive LLMs in high-stakes sectors. Contributions include an open-source evaluation notebook and filtered dataset subset, advancing reproducible aviation AI research.
VirgilAI, 2025
Air travel during pregnancy poses unique physiological risks, including deep vein thrombosis (DVT... more Air travel during pregnancy poses unique physiological risks, including deep vein thrombosis (DVT), radiation exposure, and cabin pressure effects on fetal oxygenation, necessitating rigorous fitness-to-fly assessments [?]. Current guidelines recommend medical clearance after 28 weeks of gestation or in cases of complications,
Thompson's Research Center, 2025
Advancements in artificial intelligence (AI) have enabled multimodal systems that process diverse... more Advancements in artificial intelligence (AI) have enabled multimodal systems that process diverse data types, such as text and images, to address complex user queries. This paper presents a novel Multimodal AI Chatbot designed to handle scientific text queries and image-based inputs, aligning with the goal of accelerating human scientific discovery. Built using Python, Flask, and pre-trained Transformer models from Hugging Face, the chatbot integrates a question-answering model (DistilBERT) for text processing and a vision-language model (BLIP) for image captioning, with model weights loaded securely using safetensors to mitigate vulnerabilities. The system features a user-friendly web interface and a RESTful API, demonstrating practical application in scientific contexts. Preliminary results show effective text-based answers and accurate image captions, with potential for scalability and domain-specific enhancements. This work contributes to the development of accessible AI tools for interdisciplinary research, offering insights into secure and efficient multimodal AI deployment.
ICTSolutions Africa, 2025
Holographic neural interfaces (HNIs) represent a paradigm shift in humanmachine interaction, enab... more Holographic neural interfaces (HNIs) represent a paradigm shift in humanmachine interaction, enabling seamless integration of augmented cognitive systems (ACS) with human neural processes. By leveraging holographic displays and braincomputer interfaces (BCIs), HNIs facilitate real-time cognitive augmentation for complex tasks in exascale computing environments. This paper presents a comprehensive framework for designing, implementing, and evaluating HNIs, with a focus on their scalability, adaptability, and ethical implications. Through a detailed literature review, a novel methodology, extensive simulations, and critical discussions, we demonstrate the potential of HNIs to enhance cognitive performance in fields such as scientific research, medical diagnostics, and autonomous systems management by 2050.
ICTSolutions Africa, 2025
Quantum neuromorphic computing represents a transformative paradigm that integrates quantum compu... more Quantum neuromorphic computing represents a transformative paradigm that integrates quantum computational principles with neuromorphic architectures to enable adaptive, exascale computing systems. This paper explores the theoretical foundations, design methodologies, and practical implications of quantum neuromorphic systems in achieving unprecedented computational scalability and adaptability. By leveraging quantum entanglement and spiking neural networks, we propose a framework for exascale systems capable of real-time learning and optimization in dynamic environments. This study presents a comprehensive analysis, including a review of existing literature, a novel methodology for hybrid quantumneuromorphic architectures, simulated results, and future directions for deployment in 2050's computational landscape.
gouPub, 2024
The prediction of brain diseases using computational frameworks has garnered significant attentio... more The prediction of brain diseases using computational frameworks has garnered significant attention in recent years. This paper presents a comprehensive and integrated system that leverages genomic, clinical, imaging, biomarker, behavioral, and environmental data to predict brain diseases. The framework utilizes advanced machine learning models, including Random Forest, Logistic Regression, and Variational Autoencoders (VAE), to provide accurate and interpretable predictions. The implementation is designed to be user-friendly, allowing for seamless interaction and data entry through a web-based interface. Our results demonstrate the potential of this multilevel approach in enhancing the prediction and understanding of brain diseases.
Diabetes is a global health concern with millions of new cases annually. Early detection of the d... more Diabetes is a global health concern with millions of new cases annually. Early detection of the disease can prevent its progression and complications. In this study, we developed a prediction model that uses diagnostic measurements to determine if a patient has diabetes. To improve the model's performance and accuracy, we explored different techniques instead of relying on a single algorithm or dataset, which may not be optimal for the input data or parameters. We employed Logistic Regression and Stacked Ensemble Technique, and two feature selection methods, using two datasets: the PIMA Indians Diabetes dataset and a dataset from Enugu State University Teaching Hospital. Our results show that ensemble methods improve accuracy and prediction compared to a single model. The highest accuracy achieved was 79% for Dataset 1, while employing the stacked ensemble model on Dataset 2 resulted in a 99% accuracy in predicting the blood sugar disease. Our study demonstrates the benefits of using multiple algorithms and ensemble techniques to develop accurate diabetes prediction models.
Cybersecurity in Nigeria, 2021
Cyber threats are now the most effective way to attack an individual, organization or country and... more Cyber threats are now the most effective way to attack an individual, organization or country and those with this malicious intent are finding ever more sophisticated ways of carrying out their activities. Many corporate and government entities are challenged with insufficiently secured infrastructure, lack of awareness, and are exposed to cyber attacks. Companies around the world maintain an enormous amount of personal data and records on their customers, as well as confidential information, making them frequent targets. Cyber-security is concerned with defending computers, servers, mobile devices, electronic systems, networks, and data from malicious attacks. There are substantial security holes in any web-based system; this research explains how the knowledge of cyber security helps individuals and government bodies provide reliable services to the public, maintain private communications, protect sensitive information as well as safeguard economic security.
ABSTRACT
This research work examined the impact of Commercial Bank Credit on Agriculture and Agri... more ABSTRACT
This research work examined the impact of Commercial Bank Credit on Agriculture and Agricultural Output in Nigeria using Macroeconomic variables (Commercial Bank Credit and Agricultural Output). The broad objective of the study is to investigate the extent to which commercial bank credit had supported agriculture and agricultural output in Nigeria. The specific objectives are: to determine the impact of commercial banks credit on agriculture and agricultural output in Nigeria, to determine the impact of interest rate, inflation and exchange rate on agriculture and agricultural output in Nigeria. The methodology adopted for the study was Ordinary Least Square (OLS). After the regression, the result shows that firstly: commercial bank credit and exchange rate has a positive significant effect on agriculture and agricultural output in Nigeria whereas; interest rate and the level of inflation have a negative effect on agricultural output, although they are both significant. Despite their individual signs and magnitudes, these variables all conform to economic theories. Secondly there is a general agreement that Nigeria agricultural sector is grossly underfunded. Finally, the share of actual expenditure, through commercial banks’ credit or central bank of Nigeria, that went to the agricultural sector compared unfavorable with the shares that went to other sectors. Based on the findings above, the researcher made the following suggestions: Government should take deliberate efforts to increase funding in agriculture to increase food production in the country; the government should intervene in the interest rate level in order for it to favour the common man in the country.
Academic excellence in advising receives little attention, and online academic advising receives ... more Academic excellence in advising receives little attention, and online academic advising receives even less consideration. This project explores the concept and need for academic e-advising, defined here as the systematic deployment of online instructional tools in an academic advising capacity. Techniques to encourage and disseminate such advising practices are considered, as well as general limitations and challenges in e-advising. The software is developed using ASP.net, HTML and CSS. The study concludes that E-advising brings academic advising to a medium convenient to students, and in doing so may improve the quality of advising and student academic success while enhancing online student retention among other benefits.
Uploads
Papers by Ekemini Thompson
This research work examined the impact of Commercial Bank Credit on Agriculture and Agricultural Output in Nigeria using Macroeconomic variables (Commercial Bank Credit and Agricultural Output). The broad objective of the study is to investigate the extent to which commercial bank credit had supported agriculture and agricultural output in Nigeria. The specific objectives are: to determine the impact of commercial banks credit on agriculture and agricultural output in Nigeria, to determine the impact of interest rate, inflation and exchange rate on agriculture and agricultural output in Nigeria. The methodology adopted for the study was Ordinary Least Square (OLS). After the regression, the result shows that firstly: commercial bank credit and exchange rate has a positive significant effect on agriculture and agricultural output in Nigeria whereas; interest rate and the level of inflation have a negative effect on agricultural output, although they are both significant. Despite their individual signs and magnitudes, these variables all conform to economic theories. Secondly there is a general agreement that Nigeria agricultural sector is grossly underfunded. Finally, the share of actual expenditure, through commercial banks’ credit or central bank of Nigeria, that went to the agricultural sector compared unfavorable with the shares that went to other sectors. Based on the findings above, the researcher made the following suggestions: Government should take deliberate efforts to increase funding in agriculture to increase food production in the country; the government should intervene in the interest rate level in order for it to favour the common man in the country.