TechWorld 2025: Big Data, Computer Science and Information Technologies

Ayodeji Adeyemo

Biography

Ayodeji S. Adeyemo is an emerging expert in data science and artificial intelligence, currently completing a Master of Science in Informatics (Data Science) at the University of Louisiana at Lafayette, where his thesis focuses on the application of convolutional neural networks for wildlife classification. With a foundational background in Geology from Obafemi Awolowo University, Nigeria, Ayodeji has transitioned seamlessly into the tech domain, gaining hands-on experience in machine learning, data analytics, and cloud computing. He has worked as a Data Science Intern at Walmart Global Tech ? AI Labs, where he developed a GenAI-powered pricing tool that delivered over $500K in annual cost savings. As an AI Research Scientist at Cleverarium LLC, he fine-tuned transformer-based models and led data pipeline automation, contributing to a peer-reviewed IEEE CAI 2025 paper. Ayodeji also served as a Graduate Assistant at the Region 4 STEM Network Center, creating analytical tools that enhanced educational outcomes. His technical toolkit includes Python, SQL, TensorFlow, LangChain, GCP, AWS, and various visualization platforms. Beyond academics and work, Ayodeji is active in leadership and community initiatives, including presidency roles in student and humanitarian organizations. He is passionate about leveraging data-driven technologies to solve complex real-world problems.

Research Interest

Sexing Orcas: Classification Using Convolution Neural Networks

Abstract

Identifying individual megafauna, such as orcas, is essential for understanding population dynamics, migration, and behavior. However, traditional methods like manual photo identification are time-consuming and resource-intensive. This study leverages artificial intelligence to automate orca sex classification, a crucial step toward the unique identification of individuals. A convolutional neural network model, enhanced through transfer learning with VGG16, was developed to classify orcas as male or female. Using datasets from organizations such as the Norwegian Orca Survey and the Center for Whale Research, the model achieved accuracy rates of 93% on training, 94% on validation, and 95% on test sets. This research demonstrates the potential of AI to revolutionize wildlife monitoring by improving efficiency and scalability. While challenges like image quality and overfitting persist, the adaptability of the approach offers promising applications for conservation across diverse species.