Comparative analysis and classification of elemental composition in folic acid using laser-induced breakdown spectroscopy coupled with machine learning

Folic acid (FA) tablets are widely used by pregnant women during the first few weeks of pregnancy. The qualitative analysis and classification of FA tablets from different companies of China and Pakistan are reported through a comparative laser-induced breakdown spectroscopy (LIBS) machine-learning framework. LIBS examination recognized many elements like calcium, magnesium, sodium, and potassium present in these tablets. Thirteen machine learning (ML) algorithms from the classification learner app were used for FA tablet classification. This research focused on Pakistani and Chinese tablet samples within the terms of classification. The usage of supervised ML classification models gives significant results with 91.7% and 91.6% accuracy by two models, efficient logistic regression and a wide neural network, respectively. It shows that ML algorithms are efficient at classifying FA tablets.