Abstract: This study addresses several technical limitations in the “Basic Chemistry Experiment” involving the determination of iron(III) sulfosalicylate complex, including background absorbance interference, inability to identify coordination ratios (n) in unknown systems, and low experimental throughput (measuring only one ligand per experiment). Methodologically, we developed a system of equations incorporating absorbance data from all components, effectively eliminating background interference and increasing ligand measurement capacity from 1 to 11 per experiment. Analytically, we implemented machine learning algorithms to construct a stability constant prediction model, facilitating determination of coordination ratios (n). The model was validated using UV-Vis spectral data, with feature importance analysis revealing key coordination factors, thereby providing empirical support for coordination field theory instruction. Pedagogically, we established an intelligent teaching platform to reduce technical barriers in digital implementation, promoting wider adoption of these improved methods. The enhanced protocol maintains analytical precision while significantly improving throughput and applicability. This inquiry-based approach effectively develops students’ data analysis skills and problem-solving capabilities for complex chemical challenges.

Key words: Digitalization and intelligence, Chemical experiment, Improvement and innovation, Iron complex