Advancing Systematic Literature Synthesis with Multi-Large Language Model Ensembles

Abstract: Systematic literature analysis is a powerful tool for identifying trends and gaps in scientific discovery. However, systematic literature analysis is a time-consuming process. Traditional approaches to literature synthesis—such as manual coding and keyword-based searches—often fall short in capturing the full thematic depth of scholarly discourse, especially in interdisciplinary fields. This talk presents a novel ensemble methodology that integrates multiple large language models (LLMs), including DeepSeekR1, LLaMA3, and QWEN, to retrieve, classify, and thematically cluster academic publications. Demonstrated through an analysis of over 100 peer-reviewed articles, the approach reveals latent thematic structures that surpass conventional keyword indexing, offering a more nuanced and scalable framework for academic inquiry. This methodological innovation is particularly relevant for researchers seeking rigorous and replicable approaches to knowledge mapping and agenda-setting in complex domains.

Bio: Dr. Abdullah Konak is a Distinguished Professor of Information Sciences and Technology at the Pennsylvania State University, Berks. Dr. Konak also teaches graduate courses in the Master of Science in Cybersecurity Analytics and Operations program at the College of Information Sciences and Technology, Penn State World Campus. Dr. Konak’s primary research focuses on modeling, analyzing, and optimizing complex systems using computational intelligence combined with probability, statistics, data sciences, and operations research. His research also involves active learning, entrepreneurship education, and the innovation mindset. Dr. Konak published numerous academic papers on a broad range of topics, including network design, system reliability, sustainability, cybersecurity, facilities design, green logistics, production management, and predictive analytics. Dr. Konak held visiting positions at Lehigh University and Cornell University, as well as at the Chinese University of Hong Kong, where he taught engineering innovation for over a decade. He has been a principal investigator in sponsored projects from the National Science Foundation, the National Security Agency, the U.S. Department of Labor, and Venture Well. He is a member of INFORMS and ASEE.