Enhanced artificial intelligence for electrochemical sensors in monitoring and removing of azo dyes and food colorant substances.

It is necessary to determine whether synthetic dyes are present in food since their excessive use has detrimental effects on human health. For the simultaneous assessment of tartrazine and Patent Blue V, a novel electrochemical sensing platform was developed. As a result, two artificial azo colorants (Tartrazine and Patent Blue V) with toxic azo groups (-NN-) and other carcinogenic aromatic ring structures were examined. With a low limit of detection of 0.06 μM, a broad linear concentration range 0.09μM to 950μM, and a respectable recovery, scanning electron microscopy (SEM) was able to reveal the excellent sensing performance of the suggested electrode for patent blue V. The electrochemical performance of an electrode can be characterized using cyclic and differential pulse voltammetry, and electrochemical impedance spectroscopy. Moreover, the classification model was created by applying binary classification assessment using enhanced artificial intelligence comprises of support vector machine (SVM) and Genetic Algorithm (GA), respectively, a support vector machine and a genetic algorithm, which was then validated using the 50 dyes test set. The best binary logistic regression model has an accuracy of 83.2% and 81.1%, respectively, while the best SVM model has an accuracy of 90.3% for the training group of samples and 81.1% for the test group (RMSE = 0.644, R2 = 0.873, C = 205.41, and = 5.992). According to the findings, Cu-BTC MOF (copper (II)-benzene-1,3,5-tricarboxylate) has a crystal structure and is tightly packed with hierarchically porous nanomaterials, with each particle's edge measuring between 20 and 37 nm. The suggested electrochemical sensor's analytical performance is suitable for foods like jellies, condiments, soft drinks and candies.