AI-Assisted Literature Analysis for the Experimental Design of Solvent Optimization in Analgesic Studies of Plectranthus amboinicusKamila López Díaz¹ and Manuel Reyes-Guzmán²¹Department of Chemistry, University of Puerto Rico, Río Piedras Campus²Faculty Advisor, Department of Physical Sciences, University of Puerto Rico, Río Piedras CampusThis undergraduate research project examines how artificial intelligence (AI) can enhance literature-based experimental design in the search for bioactive and analgesic compounds from Plectranthus amboinicus (Lour.) Spreng. Using AI-driven text mining and systematic review methods, published studies were analyzed to identify solvent systems and extraction parameters most frequently associated with high yields of antioxidant and phenolic constituents linked to analgesic potential.Computational clustering and bibliometric synthesis revealed consistent associations between solvent polarity and extract efficacy. Polar solvents—especially aqueous ethanol (20–40%) and methanol—were most often reported in studies describing elevated recovery of bioactive compounds. Reported statistical frameworks, including response surface methodology (RSM) and simplex–centroid optimization, showed strong predictive reliability (R² > 0.9) in previous studies, suggesting their suitability for guiding solvent optimization in future work.This phase focuses on AI-assisted literature synthesis, providing a predictive foundation for experimental design. The findings aim to support upcoming in vitro studies evaluating extraction efficiency, total phenolic content, and potential analgesic activity. By integrating computational intelligence with systematic review, this project illustrates how AI-guided data analysis can accelerate hypothesis generation, optimize research design, and deepen undergraduate engagement in phytochemical discovery.