A1227
Title: Exploring scientific literature using topic modeling: A practical framework for discovery and classification
Authors: Larry Tang - University of Central Florida (United States) [presenting]
Amir Alipour Yengejeh - University of Central Florida (United States)
Abstract: The increasing volume and diversity of scientific publications poses challenges for scalable and interpretable topic discovery and automated document categorization. An integrated framework is proposed that combines probabilistic topic modeling with supervised classification to support large-scale scientific literature analysis. Using 3689 abstracts from the Journal of Forensic Sciences, Latent Dirichlet Allocation (LDA) is applied to uncover latent thematic structures, assess topic diagnosticity across forensic disciplines, and analyze temporal research trends. The resulting document topic representations are then used for supervised abstract classification. Across multiple models and resampling scenarios, the strongest and most stable performance is achieved under a grouped category configuration.
