Çalışkan, GülizarKumluca Topallı, Ayca2024-06-292024-06-2920242470-15562470-1564https://doi.org/10.1080/24701556.2024.2354927https://hdl.handle.net/20.500.14365/5372In order to solve the modeling issues due to data scarcity problems in the disciplines utilizing statistical approximations, a novel two-stage idea is proposed. As a use case, nanoparticle biosynthesis was selected, for which an environmentally friendly process is of vital importance. First, Box Behnken Design was used for experimental setup, quadratic model formulation and data generation. The second stage consists of Machine Learning, in which the data generated in the previous stage were fed into a Neural Network to determine the relationship between the parameters. Obtained results showed that the proposed combined strategy provided better nanoparticle size estimations than the statistical approach alone. In the absence of publicly available databases, data generation using experimental design and machine learning, as proposed here, could be a faster, lower-cost, and greener solution. Our proposed method can be applied to a wide range of biotechnology and bioengineering applications with significant advanced knowledge.eninfo:eu-repo/semantics/closedAccessResponse surface methodologydata generationdeep learningneural networkssilver nanoparticle biosynthesisResponse-Surface MethodologyBiological SynthesisPredictionArticle10.1080/24701556.2024.23549272-s2.0-85193928536