Publications
Explore my research and publications in the field of data science.
Advancing clinical trial outcomes using deep learning and predictive modelling: Bridging precision medicine and patient-centered care
arXiv, 2024
DOI: 10.48550/arXiv.2412.07050
This study explores how deep learning and predictive modeling enhance clinical trials by improving patient stratification, outcome prediction, and personalized treatment. It demonstrates AI’s potential to streamline workflows, reduce costs, and advance precision medicine.
An Empirical Study of Causal Relation Extraction Transfer: Design and Data
arXiv, 2025
DOI: 10.48550/arXiv.2503.06076
This study analyzes neural architectures and data transfer strategies for causal relation extraction, finding BioBERT-BiGRU performs best across diverse sources. A new noun phrase-focused metric, 𝐹1𝑝ℎ𝑟𝑎𝑠𝑒, shows that domain-diverse data augmentation improves transfer performance.
Assessment Of Barriers to The Utilization of Primary Healthcare Services in Abuja, Nigeria.
Research Square, 2024
DOI: 10.21203/rs.3.rs-5645347/v1
This study examined barriers to primary healthcare use in Bwari, Abuja, identifying long wait times, poor facilities, and stigma as key challenges. Improving infrastructure, reducing costs, and raising awareness are recommended to boost access and satisfaction.
Can Deep Learning Large Language Models be Used to Unravel Knowledge Graph Creation?
Association for Computing Machinery, 2024
DOI: 10.1145/3661725.3661733
The paper evaluates SpaCy, BERT, Bio-BERT, and ELECTRA for extracting medical relationships from GERD-related texts. Bio-BERT outperforms others in building a reliable knowledge graph due to its biomedical training.
Understanding Graph Databases: A Comprehensive Tutorial and Survey
arXiv, 2024
DOI: 10.48550/arXiv:2411.09999
This paper offers a comprehensive guide to graph databases, covering graph theory basics, key algorithms, practical tools like Neo4j and NetworkX, and real-world applications across various domains.
Data-To-Question Generation Using Deep Learning
IEEE, 2023
DOI: 10.1109/IBDAP58581.2023.10271940
This paper presents a deep learning-based system that generates factual questions from relational datasets through a five-step pipeline. The approach often produces more accurate, data-grounded questions than general AI models like ChatGPT.